bibtex_url
stringlengths 41
53
| proceedings
stringlengths 38
50
| bibtext
stringlengths 566
3.75k
| abstract
stringlengths 4
3.1k
| authors
sequencelengths 1
66
| title
stringlengths 12
172
| id
stringlengths 7
19
| type
stringclasses 2
values | arxiv_id
stringlengths 0
10
| GitHub
sequencelengths 1
1
| paper_page
stringlengths 0
40
| n_linked_authors
int64 -1
21
| upvotes
int64 -1
116
| num_comments
int64 -1
11
| n_authors
int64 -1
61
| Models
sequencelengths 0
100
| Datasets
sequencelengths 0
100
| Spaces
sequencelengths 0
100
| old_Models
sequencelengths 0
100
| old_Datasets
sequencelengths 0
100
| old_Spaces
sequencelengths 0
100
| paper_page_exists_pre_conf
int64 0
1
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
https://aclanthology.org/2024.emnlp-main.701.bib | https://aclanthology.org/2024.emnlp-main.701/ | @inproceedings{dao-etal-2024-lifelong,
title = "Lifelong Event Detection via Optimal Transport",
author = "Dao, Viet and
Pham, Van-Cuong and
Tran, Quyen and
Le, Thanh-Thien and
Ngo, Linh Van and
Nguyen, Thien Huu",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.701",
pages = "12610--12621",
abstract = "Continual Event Detection (CED) poses a formidable challenge due to the catastrophic forgetting phenomenon, where learning new tasks (with new coming event types) hampers performance on previous ones. In this paper, we introduce a novel approach, Lifelong Event Detection via Optimal Transport (**LEDOT**), that leverages optimal transport principles to align the optimization of our classification module with the intrinsic nature of each class, as defined by their pre-trained language modeling. Our method integrates replay sets, prototype latent representations, and an innovative Optimal Transport component. Extensive experiments on MAVEN and ACE datasets demonstrate LEDOT{'}s superior performance, consistently outperforming state-of-the-art baselines. The results underscore LEDOT as a pioneering solution in continual event detection, offering a more effective and nuanced approach to addressing catastrophic forgetting in evolving environments.",
}
| Continual Event Detection (CED) poses a formidable challenge due to the catastrophic forgetting phenomenon, where learning new tasks (with new coming event types) hampers performance on previous ones. In this paper, we introduce a novel approach, Lifelong Event Detection via Optimal Transport (**LEDOT**), that leverages optimal transport principles to align the optimization of our classification module with the intrinsic nature of each class, as defined by their pre-trained language modeling. Our method integrates replay sets, prototype latent representations, and an innovative Optimal Transport component. Extensive experiments on MAVEN and ACE datasets demonstrate LEDOT{'}s superior performance, consistently outperforming state-of-the-art baselines. The results underscore LEDOT as a pioneering solution in continual event detection, offering a more effective and nuanced approach to addressing catastrophic forgetting in evolving environments. | [
"Dao, Viet",
"Pham, Van-Cuong",
"Tran, Quyen",
"Le, Thanh-Thien",
"Ngo, Linh Van",
"Nguyen, Thien Huu"
] | Lifelong Event Detection via Optimal Transport | emnlp-main.701 | Poster | 2410.08905 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.702.bib | https://aclanthology.org/2024.emnlp-main.702/ | @inproceedings{bogin-etal-2024-super,
title = "{SUPER}: Evaluating Agents on Setting Up and Executing Tasks from Research Repositories",
author = "Bogin, Ben and
Yang, Kejuan and
Gupta, Shashank and
Richardson, Kyle and
Bransom, Erin and
Clark, Peter and
Sabharwal, Ashish and
Khot, Tushar",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.702",
pages = "12622--12645",
abstract = "Given that Large Language Models (LLMs) have made significant progress in writing code, can they now be used to autonomously reproduce results from research repositories? Such a capability would be a boon to the research community, helping researchers validate, understand, and extend prior work. To advance towards this goal, we introduce SUPER, the first benchmark designed to evaluate the capability of LLMs in setting up and executing tasks from research repositories. SUPER aims to capture the realistic challenges faced by researchers working with Machine Learning (ML) and Natural Language Processing (NLP) research repositories. Our benchmark comprises three distinct problem sets: 45 end-to-end problems with annotated expert solutions, 152 sub-problems derived from the expert set that focus on specific challenges (e.g., configuring a trainer), and 602 automatically generated problems for larger-scale development. We introduce various evaluation measures to assess both task success and progress, utilizing gold solutions when available or approximations otherwise. We show that state-of-the-art approaches struggle to solve these problems with the best model (GPT-4o) solving only 16.3{\%} of the end-to-end set, and 46.1{\%} of the scenarios. This illustrates the challenge of this task, and suggests that SUPER can serve as a valuable resource for the community to make and measure progress.",
}
| Given that Large Language Models (LLMs) have made significant progress in writing code, can they now be used to autonomously reproduce results from research repositories? Such a capability would be a boon to the research community, helping researchers validate, understand, and extend prior work. To advance towards this goal, we introduce SUPER, the first benchmark designed to evaluate the capability of LLMs in setting up and executing tasks from research repositories. SUPER aims to capture the realistic challenges faced by researchers working with Machine Learning (ML) and Natural Language Processing (NLP) research repositories. Our benchmark comprises three distinct problem sets: 45 end-to-end problems with annotated expert solutions, 152 sub-problems derived from the expert set that focus on specific challenges (e.g., configuring a trainer), and 602 automatically generated problems for larger-scale development. We introduce various evaluation measures to assess both task success and progress, utilizing gold solutions when available or approximations otherwise. We show that state-of-the-art approaches struggle to solve these problems with the best model (GPT-4o) solving only 16.3{\%} of the end-to-end set, and 46.1{\%} of the scenarios. This illustrates the challenge of this task, and suggests that SUPER can serve as a valuable resource for the community to make and measure progress. | [
"Bogin, Ben",
"Yang, Kejuan",
"Gupta, Shashank",
"Richardson, Kyle",
"Bransom, Erin",
"Clark, Peter",
"Sabharwal, Ashish",
"Khot, Tushar"
] | SUPER: Evaluating Agents on Setting Up and Executing Tasks from Research Repositories | emnlp-main.702 | Poster | 2409.07440 | [
"https://github.com/allenai/super-benchmark"
] | https://huggingface.co/papers/2409.07440 | 3 | 6 | 2 | 8 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.703.bib | https://aclanthology.org/2024.emnlp-main.703/ | @inproceedings{shum-etal-2024-first,
title = "{FIRST}: Teach A Reliable Large Language Model Through Efficient Trustworthy Distillation",
author = "Shum, KaShun and
Xu, Minrui and
Zhang, Jianshu and
Chen, Zixin and
Diao, Shizhe and
Dong, Hanze and
Zhang, Jipeng and
Raza, Muhammad Omer",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.703",
pages = "12646--12659",
abstract = "Large language models (LLMs) have become increasingly prevalent in our daily lives, leading to an expectation for LLMs to be trustworthy {---}- both accurate and well-calibrated (the prediction confidence should align with its ground truth correctness likelihood). Nowadays, fine-tuning has become the most popular method for adapting a model to practical usage by significantly increasing accuracy on downstream tasks. Despite the great accuracy it achieves, we found fine-tuning is still far away from satisfactory trustworthiness due to {``}tuning-induced mis-calibration{''}. In this paper, we delve deeply into why and how mis-calibration exists in fine-tuned models, and how distillation can alleviate the issue. Then we further propose a brand new method named Efficient Trustworthy Distillation (FIRST), which utilizes a small portion of teacher{'}s knowledge to obtain a reliable language model in a cost-efficient way. Specifically, we identify the {``}concentrated knowledge{''} phenomenon during distillation, which can significantly reduce the computational burden. Then we apply a {``}trustworthy maximization{''} process to optimize the utilization of this small portion of concentrated knowledge before transferring it to the student. Experimental results demonstrate the effectiveness of our method, where better accuracy (+2.3{\%}) and less mis-calibration (-10{\%}) are achieved on average across both in-domain and out-of-domain scenarios, indicating better trustworthiness.",
}
| Large language models (LLMs) have become increasingly prevalent in our daily lives, leading to an expectation for LLMs to be trustworthy {---}- both accurate and well-calibrated (the prediction confidence should align with its ground truth correctness likelihood). Nowadays, fine-tuning has become the most popular method for adapting a model to practical usage by significantly increasing accuracy on downstream tasks. Despite the great accuracy it achieves, we found fine-tuning is still far away from satisfactory trustworthiness due to {``}tuning-induced mis-calibration{''}. In this paper, we delve deeply into why and how mis-calibration exists in fine-tuned models, and how distillation can alleviate the issue. Then we further propose a brand new method named Efficient Trustworthy Distillation (FIRST), which utilizes a small portion of teacher{'}s knowledge to obtain a reliable language model in a cost-efficient way. Specifically, we identify the {``}concentrated knowledge{''} phenomenon during distillation, which can significantly reduce the computational burden. Then we apply a {``}trustworthy maximization{''} process to optimize the utilization of this small portion of concentrated knowledge before transferring it to the student. Experimental results demonstrate the effectiveness of our method, where better accuracy (+2.3{\%}) and less mis-calibration (-10{\%}) are achieved on average across both in-domain and out-of-domain scenarios, indicating better trustworthiness. | [
"Shum, KaShun",
"Xu, Minrui",
"Zhang, Jianshu",
"Chen, Zixin",
"Diao, Shizhe",
"Dong, Hanze",
"Zhang, Jipeng",
"Raza, Muhammad Omer"
] | FIRST: Teach A Reliable Large Language Model Through Efficient Trustworthy Distillation | emnlp-main.703 | Poster | 2408.12168 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.704.bib | https://aclanthology.org/2024.emnlp-main.704/ | @inproceedings{saunders-deneefe-2024-domain,
title = "Domain adapted machine translation: What does catastrophic forgetting forget and why?",
author = "Saunders, Danielle and
DeNeefe, Steve",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.704",
pages = "12660--12671",
abstract = "Neural Machine Translation (NMT) models can be specialized by domain adaptation, often involving fine-tuning on a dataset of interest. This process risks catastrophic forgetting: rapid loss of generic translation quality. Forgetting has been widely observed, with many mitigation methods proposed. However, the causes of forgetting and the relationship between forgetting and adaptation data are underexplored.This paper takes a novel approach to understanding catastrophic forgetting during NMT adaptation by investigating the impact of the data. We provide a first investigation of what is forgotten, and why. We examine the relationship between forgetting and the in-domain data, and show that the amount and type of forgetting is linked to that data{'}s target vocabulary coverage. Our findings pave the way toward better informed NMT domain adaptation.",
}
| Neural Machine Translation (NMT) models can be specialized by domain adaptation, often involving fine-tuning on a dataset of interest. This process risks catastrophic forgetting: rapid loss of generic translation quality. Forgetting has been widely observed, with many mitigation methods proposed. However, the causes of forgetting and the relationship between forgetting and adaptation data are underexplored.This paper takes a novel approach to understanding catastrophic forgetting during NMT adaptation by investigating the impact of the data. We provide a first investigation of what is forgotten, and why. We examine the relationship between forgetting and the in-domain data, and show that the amount and type of forgetting is linked to that data{'}s target vocabulary coverage. Our findings pave the way toward better informed NMT domain adaptation. | [
"Saunders, Danielle",
"DeNeefe, Steve"
] | Domain adapted machine translation: What does catastrophic forgetting forget and why? | emnlp-main.704 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.705.bib | https://aclanthology.org/2024.emnlp-main.705/ | @inproceedings{towle-zhou-2024-enhancing,
title = "Enhancing {AI} Assisted Writing with One-Shot Implicit Negative Feedback",
author = "Towle, Benjamin and
Zhou, Ke",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.705",
pages = "12672--12680",
abstract = "AI-mediated communication enables users to communicate more quickly and efficiently. Various systems have been proposed such as smart reply and AI-assisted writing. Yet, the heterogeneity of the forms of inputs and architectures often renders it challenging to combine insights from user behaviour in one system to improve performance in another. In this work, we consider the case where the user does not select any of the suggested replies from a smart reply system, and how this can be used as one-shot implicit negative feedback to enhance the accuracy of an AI writing model. We introduce Nifty, an approach that uses classifier guidance to controllably integrate implicit user feedback into the text generation process. Empirically, we find up to 34{\%} improvement in Rouge-L, 89{\%} improvement in generating the correct intent, and an 86{\%} win-rate according to human evaluators compared to a vanilla AI writing system on the MultiWOZ and Schema-Guided Dialog datasets. The code is available at https://github.com/BenjaminTowle/NIFTY.",
}
| AI-mediated communication enables users to communicate more quickly and efficiently. Various systems have been proposed such as smart reply and AI-assisted writing. Yet, the heterogeneity of the forms of inputs and architectures often renders it challenging to combine insights from user behaviour in one system to improve performance in another. In this work, we consider the case where the user does not select any of the suggested replies from a smart reply system, and how this can be used as one-shot implicit negative feedback to enhance the accuracy of an AI writing model. We introduce Nifty, an approach that uses classifier guidance to controllably integrate implicit user feedback into the text generation process. Empirically, we find up to 34{\%} improvement in Rouge-L, 89{\%} improvement in generating the correct intent, and an 86{\%} win-rate according to human evaluators compared to a vanilla AI writing system on the MultiWOZ and Schema-Guided Dialog datasets. The code is available at https://github.com/BenjaminTowle/NIFTY. | [
"Towle, Benjamin",
"Zhou, Ke"
] | Enhancing AI Assisted Writing with One-Shot Implicit Negative Feedback | emnlp-main.705 | Poster | 2410.11009 | [
"https://github.com/BenjaminTowle/NIFTY"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.706.bib | https://aclanthology.org/2024.emnlp-main.706/ | @inproceedings{thirukovalluru-etal-2024-atomic,
title = "Atomic Self-Consistency for Better Long Form Generations",
author = "Thirukovalluru, Raghuveer and
Huang, Yukun and
Dhingra, Bhuwan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.706",
pages = "12681--12694",
abstract = "Recent work has aimed to improve LLM generations by filtering out hallucinations, thereby improving the precision of the information in responses. Correctness of a long-form response, however, also depends on the recall of multiple pieces of information relevant to the question. In this paper, we introduce Atomic Self-Consistency (ASC), a technique for improving the recall of relevant information in an LLM response. ASC follows recent work, Universal Self-Consistency (USC) in using multiple stochastic samples from an LLM to improve the long-form response. Unlike USC which only focuses on selecting the best single generation, ASC picks authentic subparts from the samples and merges them into a superior composite answer. Through extensive experiments and ablations, we show that merging relevant subparts of multiple samples performs significantly better than picking a single sample. ASC demonstrates significant gains over USC on multiple factoids and open-ended QA datasets - ASQA, QAMPARI, QUEST, ELI5 with ChatGPT and Llama3. Our analysis also reveals untapped potential for enhancing long-form generations using the approach of merging multiple samples.",
}
| Recent work has aimed to improve LLM generations by filtering out hallucinations, thereby improving the precision of the information in responses. Correctness of a long-form response, however, also depends on the recall of multiple pieces of information relevant to the question. In this paper, we introduce Atomic Self-Consistency (ASC), a technique for improving the recall of relevant information in an LLM response. ASC follows recent work, Universal Self-Consistency (USC) in using multiple stochastic samples from an LLM to improve the long-form response. Unlike USC which only focuses on selecting the best single generation, ASC picks authentic subparts from the samples and merges them into a superior composite answer. Through extensive experiments and ablations, we show that merging relevant subparts of multiple samples performs significantly better than picking a single sample. ASC demonstrates significant gains over USC on multiple factoids and open-ended QA datasets - ASQA, QAMPARI, QUEST, ELI5 with ChatGPT and Llama3. Our analysis also reveals untapped potential for enhancing long-form generations using the approach of merging multiple samples. | [
"Thirukovalluru, Raghuveer",
"Huang, Yukun",
"Dhingra, Bhuwan"
] | Atomic Self-Consistency for Better Long Form Generations | emnlp-main.706 | Poster | 2405.13131 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.707.bib | https://aclanthology.org/2024.emnlp-main.707/ | @inproceedings{kamruzzaman-etal-2024-global,
title = "{``}Global is Good, Local is Bad?{''}: Understanding Brand Bias in {LLM}s",
author = "Kamruzzaman, Mahammed and
Nguyen, Hieu Minh and
Kim, Gene Louis",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.707",
pages = "12695--12702",
abstract = "Many recent studies have investigated social biases in LLMs but brand bias has received little attention. This research examines the biases exhibited by LLMs towards different brands, a significant concern given the widespread use of LLMs in affected use cases such as product recommendation and market analysis. Biased models may perpetuate societal inequalities, unfairly favoring established global brands while marginalizing local ones. Using a curated dataset across four brand categories, we probe the behavior of LLMs in this space. We find a consistent pattern of bias in this space{---}both in terms of disproportionately associating global brands with positive attributes and disproportionately recommending luxury gifts for individuals in high-income countries. We also find LLMs are subject to country-of-origin effects which may boost local brand preference in LLM outputs in specific contexts.",
}
| Many recent studies have investigated social biases in LLMs but brand bias has received little attention. This research examines the biases exhibited by LLMs towards different brands, a significant concern given the widespread use of LLMs in affected use cases such as product recommendation and market analysis. Biased models may perpetuate societal inequalities, unfairly favoring established global brands while marginalizing local ones. Using a curated dataset across four brand categories, we probe the behavior of LLMs in this space. We find a consistent pattern of bias in this space{---}both in terms of disproportionately associating global brands with positive attributes and disproportionately recommending luxury gifts for individuals in high-income countries. We also find LLMs are subject to country-of-origin effects which may boost local brand preference in LLM outputs in specific contexts. | [
"Kamruzzaman, Mahammed",
"Nguyen, Hieu Minh",
"Kim, Gene Louis"
] | “Global is Good, Local is Bad?”: Understanding Brand Bias in LLMs | emnlp-main.707 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.708.bib | https://aclanthology.org/2024.emnlp-main.708/ | @inproceedings{li-etal-2024-optimizing-rare,
title = "Optimizing Rare Word Accuracy in Direct Speech Translation with a Retrieval-and-Demonstration Approach",
author = "Li, Siqi and
Liu, Danni and
Niehues, Jan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.708",
pages = "12703--12719",
abstract = "Direct speech translation (ST) models often struggle with rare words. Incorrect translation of these words can have severe consequences, impacting translation quality and user trust. While rare word translation is inherently challenging for neural models due to sparse learning signals, real-world scenarios often allow access to translations of past recordings on similar topics. To leverage these valuable resources, we propose a retrieval-and-demonstration approach to enhance rare word translation accuracy in direct ST models. First, we adapt existing ST models to incorporate retrieved examples for rare word translation, which allows the model to benefit from prepended examples, similar to in-context learning. We then develop a cross-modal (speech-to-speech, speech-to-text, text-to-text) retriever to locate suitable examples. We demonstrate that standard ST models can be effectively adapted to leverage examples for rare word translation, improving rare word translation accuracy over the baseline by 17.6{\%} with gold examples and 8.5{\%} with retrieved examples. Moreover, our speech-to-speech retrieval approach outperforms other modalities and exhibits higher robustness to unseen speakers. Our code is publicly available.",
}
| Direct speech translation (ST) models often struggle with rare words. Incorrect translation of these words can have severe consequences, impacting translation quality and user trust. While rare word translation is inherently challenging for neural models due to sparse learning signals, real-world scenarios often allow access to translations of past recordings on similar topics. To leverage these valuable resources, we propose a retrieval-and-demonstration approach to enhance rare word translation accuracy in direct ST models. First, we adapt existing ST models to incorporate retrieved examples for rare word translation, which allows the model to benefit from prepended examples, similar to in-context learning. We then develop a cross-modal (speech-to-speech, speech-to-text, text-to-text) retriever to locate suitable examples. We demonstrate that standard ST models can be effectively adapted to leverage examples for rare word translation, improving rare word translation accuracy over the baseline by 17.6{\%} with gold examples and 8.5{\%} with retrieved examples. Moreover, our speech-to-speech retrieval approach outperforms other modalities and exhibits higher robustness to unseen speakers. Our code is publicly available. | [
"Li, Siqi",
"Liu, Danni",
"Niehues, Jan"
] | Optimizing Rare Word Accuracy in Direct Speech Translation with a Retrieval-and-Demonstration Approach | emnlp-main.708 | Poster | 2409.09009 | [
"https://github.com/siqilii/retrieve-and-demonstration-st"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.709.bib | https://aclanthology.org/2024.emnlp-main.709/ | @inproceedings{shea-etal-2024-ace,
title = "{ACE}: A {LLM}-based Negotiation Coaching System",
author = "Shea, Ryan and
Kallala, Aymen and
Liu, Xin Lucy and
Morris, Michael W. and
Yu, Zhou",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.709",
pages = "12720--12749",
abstract = "The growing prominence of LLMs has led to an increase in the development of AI tutoring systems. These systems are crucial in providing underrepresented populations with improved access to valuable education. One important area of education that is unavailable to many learners is strategic bargaining related to negotiation. To address this, we develop a LLM-based Assistant for Coaching nEgotiation (ACE). ACE not only serves as a negotiation partner for users but also provides them with targeted feedback for improvement. To build our system, we collect a dataset of negotiation transcripts between MBA students. These transcripts come from trained negotiators and emulate realistic bargaining scenarios. We use the dataset, along with expert consultations, to design an annotation scheme for detecting negotiation mistakes. ACE employs this scheme to identify mistakes and provide targeted feedback to users. To test the effectiveness of ACE-generated feedback, we conducted a user experiment with two consecutive trials of negotiation and found that it improves negotiation performances significantly compared to a system that doesn{'}t provide feedback and one which uses an alternative method of providing feedback.",
}
| The growing prominence of LLMs has led to an increase in the development of AI tutoring systems. These systems are crucial in providing underrepresented populations with improved access to valuable education. One important area of education that is unavailable to many learners is strategic bargaining related to negotiation. To address this, we develop a LLM-based Assistant for Coaching nEgotiation (ACE). ACE not only serves as a negotiation partner for users but also provides them with targeted feedback for improvement. To build our system, we collect a dataset of negotiation transcripts between MBA students. These transcripts come from trained negotiators and emulate realistic bargaining scenarios. We use the dataset, along with expert consultations, to design an annotation scheme for detecting negotiation mistakes. ACE employs this scheme to identify mistakes and provide targeted feedback to users. To test the effectiveness of ACE-generated feedback, we conducted a user experiment with two consecutive trials of negotiation and found that it improves negotiation performances significantly compared to a system that doesn{'}t provide feedback and one which uses an alternative method of providing feedback. | [
"Shea, Ryan",
"Kallala, Aymen",
"Liu, Xin Lucy",
"Morris, Michael W.",
"Yu, Zhou"
] | ACE: A LLM-based Negotiation Coaching System | emnlp-main.709 | Oral | 2410.01555 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.710.bib | https://aclanthology.org/2024.emnlp-main.710/ | @inproceedings{zhang-etal-2024-transfertod,
title = "{T}ransfer{TOD}: A Generalizable {C}hinese Multi-Domain Task-Oriented Dialogue System with Transfer Capabilities",
author = "Zhang, Ming and
Huang, Caishuang and
Wu, Yilong and
Liu, Shichun and
Zheng, Huiyuan and
Dong, Yurui and
Shen, Yujiong and
Dou, Shihan and
Zhao, Jun and
Ye, Junjie and
Zhang, Qi and
Gui, Tao and
Huang, Xuanjing",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.710",
pages = "12750--12771",
abstract = "Task-oriented dialogue (TOD) systems aim to efficiently handle task-oriented conversations, including information collection. How to utilize TOD accurately, efficiently and effectively for information collection has always been a critical and challenging task. Recent studies have demonstrated that Large Language Models (LLMs) excel in dialogue, instruction generation, and reasoning, and can significantly enhance the performance of TOD through fine-tuning. However, current datasets primarily cater to user-led systems and are limited to predefined specific scenarios and slots, thereby necessitating improvements in the proactiveness, diversity, and capabilities of TOD. In this study, we present a detailed multi-domain task-oriented data construction process for conversations, and a Chinese dialogue dataset generated based on this process, **TransferTOD**, which authentically simulates human-computer dialogues in 30 popular life service scenarios. Leveraging this dataset, we trained a model using full-parameter fine-tuning called **TransferTOD-7B**, showcasing notable abilities in slot filling and questioning. Our work has demonstrated its strong generalization capabilities in various downstream scenarios, significantly enhancing both data utilization efficiency and system performance. The data is released in https://github.com/KongLongGeFDU/TransferTOD.",
}
| Task-oriented dialogue (TOD) systems aim to efficiently handle task-oriented conversations, including information collection. How to utilize TOD accurately, efficiently and effectively for information collection has always been a critical and challenging task. Recent studies have demonstrated that Large Language Models (LLMs) excel in dialogue, instruction generation, and reasoning, and can significantly enhance the performance of TOD through fine-tuning. However, current datasets primarily cater to user-led systems and are limited to predefined specific scenarios and slots, thereby necessitating improvements in the proactiveness, diversity, and capabilities of TOD. In this study, we present a detailed multi-domain task-oriented data construction process for conversations, and a Chinese dialogue dataset generated based on this process, **TransferTOD**, which authentically simulates human-computer dialogues in 30 popular life service scenarios. Leveraging this dataset, we trained a model using full-parameter fine-tuning called **TransferTOD-7B**, showcasing notable abilities in slot filling and questioning. Our work has demonstrated its strong generalization capabilities in various downstream scenarios, significantly enhancing both data utilization efficiency and system performance. The data is released in https://github.com/KongLongGeFDU/TransferTOD. | [
"Zhang, Ming",
"Huang, Caishuang",
"Wu, Yilong",
"Liu, Shichun",
"Zheng, Huiyuan",
"Dong, Yurui",
"Shen, Yujiong",
"Dou, Shihan",
"Zhao, Jun",
"Ye, Junjie",
"Zhang, Qi",
"Gui, Tao",
"Huang, Xuanjing"
] | TransferTOD: A Generalizable Chinese Multi-Domain Task-Oriented Dialogue System with Transfer Capabilities | emnlp-main.710 | Poster | 2407.21693 | [
"https://github.com/konglonggefdu/transfertod"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.711.bib | https://aclanthology.org/2024.emnlp-main.711/ | @inproceedings{wang-etal-2024-patient,
title = "{PATIENT}-$\psi$: Using Large Language Models to Simulate Patients for Training Mental Health Professionals",
author = "Wang, Ruiyi and
Milani, Stephanie and
Chiu, Jamie C. and
Zhi, Jiayin and
Eack, Shaun M. and
Labrum, Travis and
Murphy, Samuel M and
Jones, Nev and
Hardy, Kate V and
Shen, Hong and
Fang, Fei and
Chen, Zhiyu",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.711",
pages = "12772--12797",
abstract = "Mental illness remains one of the most critical public health issues. Despite its importance, many mental health professionals highlight a disconnect between their training and actual real-world patient practice. To help bridge this gap, we propose PATIENT-$\psi$, a novel patient simulation framework for cognitive behavior therapy (CBT) training. To build PATIENT-$\psi$, we construct diverse patient cognitive models based on CBT principles and use large language models (LLMs) programmed with these cognitive models to act as a simulated therapy patient. We propose an interactive training scheme, PATIENT-$\psi$-TRAINER, for mental health trainees to practice a key skill in CBT {--} formulating the cognitive model of the patient {--} through role-playing a therapy session with PATIENT-$\psi$. To evaluate PATIENT-$\psi$, we conducted a comprehensive user study of 13 mental health trainees and 20 experts. The results demonstrate that practice using PATIENT-$\psi$-TRAINER enhances the perceived skill acquisition and confidence of the trainees beyond existing forms of training such as textbooks, videos, and role-play with non-patients. Based on the experts{'} perceptions, PATIENT-$\psi$ is perceived to be closer to real patient interactions than GPT-4, and PATIENT-$\psi$-TRAINER holds strong promise to improve trainee competencies. Our code and data are released at \url{https://github.com/ruiyiw/patient-psi}.",
}
| Mental illness remains one of the most critical public health issues. Despite its importance, many mental health professionals highlight a disconnect between their training and actual real-world patient practice. To help bridge this gap, we propose PATIENT-$\psi$, a novel patient simulation framework for cognitive behavior therapy (CBT) training. To build PATIENT-$\psi$, we construct diverse patient cognitive models based on CBT principles and use large language models (LLMs) programmed with these cognitive models to act as a simulated therapy patient. We propose an interactive training scheme, PATIENT-$\psi$-TRAINER, for mental health trainees to practice a key skill in CBT {--} formulating the cognitive model of the patient {--} through role-playing a therapy session with PATIENT-$\psi$. To evaluate PATIENT-$\psi$, we conducted a comprehensive user study of 13 mental health trainees and 20 experts. The results demonstrate that practice using PATIENT-$\psi$-TRAINER enhances the perceived skill acquisition and confidence of the trainees beyond existing forms of training such as textbooks, videos, and role-play with non-patients. Based on the experts{'} perceptions, PATIENT-$\psi$ is perceived to be closer to real patient interactions than GPT-4, and PATIENT-$\psi$-TRAINER holds strong promise to improve trainee competencies. Our code and data are released at \url{https://github.com/ruiyiw/patient-psi}. | [
"Wang, Ruiyi",
"Milani, Stephanie",
"Chiu, Jamie C.",
"Zhi, Jiayin",
"Eack, Shaun M.",
"Labrum, Travis",
"Murphy, Samuel M",
"Jones, Nev",
"Hardy, Kate V",
"Shen, Hong",
"Fang, Fei",
"Chen, Zhiyu"
] | PATIENT-ψ: Using Large Language Models to Simulate Patients for Training Mental Health Professionals | emnlp-main.711 | Poster | 2405.19660 | [
"https://github.com/ruiyiw/patient-psi"
] | https://huggingface.co/papers/2405.19660 | 0 | 0 | 0 | 12 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.712.bib | https://aclanthology.org/2024.emnlp-main.712/ | @inproceedings{ge-etal-2024-dkec,
title = "{DKEC}: Domain Knowledge Enhanced Multi-Label Classification for Diagnosis Prediction",
author = "Ge, Xueren and
Satpathy, Abhishek and
Williams, Ronald Dean and
Stankovic, John and
Alemzadeh, Homa",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.712",
pages = "12798--12813",
abstract = "Multi-label text classification (MLTC) tasks in the medical domain often face the long-tail label distribution problem. Prior works have explored hierarchical label structures to find relevant information for few-shot classes, but mostly neglected to incorporate external knowledge from medical guidelines. This paper presents DKEC, Domain Knowledge Enhanced Classification for diagnosis prediction with two innovations: (1) automated construction of heterogeneous knowledge graphs from external sources to capture semantic relations among diverse medical entities, (2) incorporating the heterogeneous knowledge graphs in few-shot classification using a label-wise attention mechanism. We construct DKEC using three online medical knowledge sources and evaluate it on a real-world Emergency Medical Services (EMS) dataset and a public electronic health record (EHR) dataset. Results show that DKEC outperforms the state-of-the-art label-wise attention networks and transformer models of different sizes, particularly for the few-shot classes. More importantly, it helps the smaller language models achieve comparable performance to large language models.",
}
| Multi-label text classification (MLTC) tasks in the medical domain often face the long-tail label distribution problem. Prior works have explored hierarchical label structures to find relevant information for few-shot classes, but mostly neglected to incorporate external knowledge from medical guidelines. This paper presents DKEC, Domain Knowledge Enhanced Classification for diagnosis prediction with two innovations: (1) automated construction of heterogeneous knowledge graphs from external sources to capture semantic relations among diverse medical entities, (2) incorporating the heterogeneous knowledge graphs in few-shot classification using a label-wise attention mechanism. We construct DKEC using three online medical knowledge sources and evaluate it on a real-world Emergency Medical Services (EMS) dataset and a public electronic health record (EHR) dataset. Results show that DKEC outperforms the state-of-the-art label-wise attention networks and transformer models of different sizes, particularly for the few-shot classes. More importantly, it helps the smaller language models achieve comparable performance to large language models. | [
"Ge, Xueren",
"Satpathy, Abhishek",
"Williams, Ronald Dean",
"Stankovic, John",
"Alemzadeh, Homa"
] | DKEC: Domain Knowledge Enhanced Multi-Label Classification for Diagnosis Prediction | emnlp-main.712 | Poster | 2310.07059 | [
"https://github.com/uva-dsa/dkec"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.713.bib | https://aclanthology.org/2024.emnlp-main.713/ | @inproceedings{jiang-etal-2024-modscan,
title = "$\texttt{ModSCAN}$: Measuring Stereotypical Bias in Large Vision-Language Models from Vision and Language Modalities",
author = "Jiang, Yukun and
Li, Zheng and
Shen, Xinyue and
Liu, Yugeng and
Backes, Michael and
Zhang, Yang",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.713",
pages = "12814--12845",
}
| No abstract found | [
"Jiang, Yukun",
"Li, Zheng",
"Shen, Xinyue",
"Liu, Yugeng",
"Backes, Michael",
"Zhang, Yang"
] | : Measuring Stereotypical Bias in Large Vision-Language Models from Vision and Language Modalities | emnlp-main.713 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.714.bib | https://aclanthology.org/2024.emnlp-main.714/ | @inproceedings{wu-etal-2024-large,
title = "Large Language Models Can Self-Correct with Key Condition Verification",
author = "Wu, Zhenyu and
Zeng, Qingkai and
Zhang, Zhihan and
Tan, Zhaoxuan and
Shen, Chao and
Jiang, Meng",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.714",
pages = "12846--12867",
abstract = "Intrinsic self-correct was a method that instructed large language models (LLMs) to verify and correct their responses without external feedback. Unfortunately, the study concluded that the LLMs could not self-correct reasoning yet. We find that a simple yet effective prompting method enhances LLM performance in identifying and correcting inaccurate answers without external feedback.That is to mask a key condition in the question, add the current response to construct a verification question, and predict the condition to verify the response. The condition can be an entity in an open-domain question or a numerical value in an arithmetic question, which requires minimal effort (via prompting) to identify. We propose an iterative verify-then-correct framework to progressively identify and correct (probably) false responses, named ProCo. We conduct experiments on three reasoning tasks. On average, ProCo, with GPT-3.5-Turbo-1106 as the backend LLM, yields $+6.8$ exact match on four open-domain question answering datasets, $+14.1$ accuracy on three arithmetic reasoning datasets, and $+9.6$ accuracy on a commonsense reasoning dataset, compared to Self-Correct.Our implementation is made publicly available at https://wzy6642.github.io/proco.github.io/.",
}
| Intrinsic self-correct was a method that instructed large language models (LLMs) to verify and correct their responses without external feedback. Unfortunately, the study concluded that the LLMs could not self-correct reasoning yet. We find that a simple yet effective prompting method enhances LLM performance in identifying and correcting inaccurate answers without external feedback.That is to mask a key condition in the question, add the current response to construct a verification question, and predict the condition to verify the response. The condition can be an entity in an open-domain question or a numerical value in an arithmetic question, which requires minimal effort (via prompting) to identify. We propose an iterative verify-then-correct framework to progressively identify and correct (probably) false responses, named ProCo. We conduct experiments on three reasoning tasks. On average, ProCo, with GPT-3.5-Turbo-1106 as the backend LLM, yields $+6.8$ exact match on four open-domain question answering datasets, $+14.1$ accuracy on three arithmetic reasoning datasets, and $+9.6$ accuracy on a commonsense reasoning dataset, compared to Self-Correct.Our implementation is made publicly available at https://wzy6642.github.io/proco.github.io/. | [
"Wu, Zhenyu",
"Zeng, Qingkai",
"Zhang, Zhihan",
"Tan, Zhaoxuan",
"Shen, Chao",
"Jiang, Meng"
] | Large Language Models Can Self-Correct with Key Condition Verification | emnlp-main.714 | Poster | 2405.14092 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.715.bib | https://aclanthology.org/2024.emnlp-main.715/ | @inproceedings{tang-van-hell-2024-learning,
title = "Learning to Write Rationally: How Information Is Distributed in Non-native Speakers{'} Essays",
author = "Tang, Zixin and
Van Hell, Janet",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.715",
pages = "12868--12879",
abstract = "People tend to distribute information evenly in language production for better and clearer communication. In this study, we compared essays written by second language (L2) learners with various native language (L1) backgrounds to investigate how they distribute information in their non-native L2 production. Analyses of surprisal and constancy of entropy rate indicated that writers with higher L2 proficiency can reduce the expected uncertainty of language production while still conveying informative content. However, the uniformity of information distribution showed less variability among different groups of L2 speakers, suggesting that this feature may be universal in L2 essay writing and less affected by L2 writers{'} variability in L1 background and L2 proficiency.",
}
| People tend to distribute information evenly in language production for better and clearer communication. In this study, we compared essays written by second language (L2) learners with various native language (L1) backgrounds to investigate how they distribute information in their non-native L2 production. Analyses of surprisal and constancy of entropy rate indicated that writers with higher L2 proficiency can reduce the expected uncertainty of language production while still conveying informative content. However, the uniformity of information distribution showed less variability among different groups of L2 speakers, suggesting that this feature may be universal in L2 essay writing and less affected by L2 writers{'} variability in L1 background and L2 proficiency. | [
"Tang, Zixin",
"Van Hell, Janet"
] | Learning to Write Rationally: How Information Is Distributed in Non-native Speakers' Essays | emnlp-main.715 | Poster | 2411.03550 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.716.bib | https://aclanthology.org/2024.emnlp-main.716/ | @inproceedings{ai-etal-2024-defending,
title = "Defending Against Social Engineering Attacks in the Age of {LLM}s",
author = "Ai, Lin and
Kumarage, Tharindu Sandaruwan and
Bhattacharjee, Amrita and
Liu, Zizhou and
Hui, Zheng and
Davinroy, Michael S. and
Cook, James and
Cassani, Laura and
Trapeznikov, Kirill and
Kirchner, Matthias and
Basharat, Arslan and
Hoogs, Anthony and
Garland, Joshua and
Liu, Huan and
Hirschberg, Julia",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.716",
pages = "12880--12902",
}
| No abstract found | [
"Ai, Lin",
"Kumarage, Tharindu S",
"aruwan",
"Bhattacharjee, Amrita",
"Liu, Zizhou",
"Hui, Zheng",
"Davinroy, Michael S.",
"Cook, James",
"Cassani, Laura",
"Trapeznikov, Kirill",
"Kirchner, Matthias",
"Basharat, Arslan",
"Hoogs, Anthony",
"Garl",
", Joshua",
"Liu, Huan",
"Hirschberg, Julia"
] | Defending Against Social Engineering Attacks in the Age of LLMs | emnlp-main.716 | Poster | 2406.12263 | [
"https://github.com/lynneeai/convosentinel"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.717.bib | https://aclanthology.org/2024.emnlp-main.717/ | @inproceedings{cho-etal-2024-heterogeneous,
title = "Heterogeneous {L}o{RA} for Federated Fine-tuning of On-Device Foundation Models",
author = "Cho, Yae Jee and
Liu, Luyang and
Xu, Zheng and
Fahrezi, Aldi and
Joshi, Gauri",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.717",
pages = "12903--12913",
abstract = "Foundation models (FMs) adapt surprisingly well to downstream tasks with fine-tuning. However, their colossal parameter space prohibits their training on resource-constrained edge-devices. For federated fine-tuning, we need to consider the smaller FMs of few billion parameters at most, namely on-device FMs (ODFMs), which can be deployed on-device. Federated fine-tuning of ODFMs has unique challenges non-present in standard fine-tuning: i) ODFMs poorly generalize to downstream tasks due to their limited sizes making proper fine-tuning imperative to their performance, and ii) devices have limited and heterogeneous system capabilities and data that can deter the performance of fine-tuning.Tackling these challenges, we propose HetLoRA, a feasible and effective federated fine-tuning method for ODFMs that leverages the system and data heterogeneity at the edge. HetLoRA allows heterogeneous LoRA ranks across clients for their individual system resources, and efficiently aggregates and distributes these LoRA modules in a data-aware manner by applying rank self-pruning locally and sparsity-weighted aggregation at the server. It combines the advantages of high and low-rank LoRAs, achieving improved convergence speed and final performance compared to homogeneous LoRA. Furthermore, HetLoRA has enhanced computation and communication efficiency compared to full fine-tuning making it more feasible for the edge.",
}
| Foundation models (FMs) adapt surprisingly well to downstream tasks with fine-tuning. However, their colossal parameter space prohibits their training on resource-constrained edge-devices. For federated fine-tuning, we need to consider the smaller FMs of few billion parameters at most, namely on-device FMs (ODFMs), which can be deployed on-device. Federated fine-tuning of ODFMs has unique challenges non-present in standard fine-tuning: i) ODFMs poorly generalize to downstream tasks due to their limited sizes making proper fine-tuning imperative to their performance, and ii) devices have limited and heterogeneous system capabilities and data that can deter the performance of fine-tuning.Tackling these challenges, we propose HetLoRA, a feasible and effective federated fine-tuning method for ODFMs that leverages the system and data heterogeneity at the edge. HetLoRA allows heterogeneous LoRA ranks across clients for their individual system resources, and efficiently aggregates and distributes these LoRA modules in a data-aware manner by applying rank self-pruning locally and sparsity-weighted aggregation at the server. It combines the advantages of high and low-rank LoRAs, achieving improved convergence speed and final performance compared to homogeneous LoRA. Furthermore, HetLoRA has enhanced computation and communication efficiency compared to full fine-tuning making it more feasible for the edge. | [
"Cho, Yae Jee",
"Liu, Luyang",
"Xu, Zheng",
"Fahrezi, Aldi",
"Joshi, Gauri"
] | Heterogeneous LoRA for Federated Fine-tuning of On-Device Foundation Models | emnlp-main.717 | Poster | 2401.06432 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.718.bib | https://aclanthology.org/2024.emnlp-main.718/ | @inproceedings{wang-etal-2024-make,
title = "Make Some Noise: Unlocking Language Model Parallel Inference Capability through Noisy Training",
author = "Wang, Yixuan and
Luo, Xianzhen and
Wei, Fuxuan and
Liu, Yijun and
Zhu, Qingfu and
Zhang, Xuanyu and
Yang, Qing and
Xu, Dongliang and
Che, Wanxiang",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.718",
pages = "12914--12926",
abstract = "Existing speculative decoding methods typically require additional model structure and training processes to assist the model for draft token generation. This makes the migration of acceleration methods to the new model more costly and more demanding on device memory. To address this problem, we propose the Make Some Noise (MSN) training framework as a replacement for the supervised fine-tuning stage of the large language model. The training method simply introduces some noise at the input for the model to learn the denoising task. It significantly enhances the parallel decoding capability of the model without affecting the original task capability. In addition, we propose a tree-based retrieval-augmented Jacobi (TR-Jacobi) decoding strategy to further improve the inference speed of MSN models. Experiments in both the general and code domains have shown that MSN can improve inference speed by 2.3-2.7x times without compromising model performance. The MSN model also achieves comparable acceleration ratios to the SOTA model with additional model structure on Spec-Bench.",
}
| Existing speculative decoding methods typically require additional model structure and training processes to assist the model for draft token generation. This makes the migration of acceleration methods to the new model more costly and more demanding on device memory. To address this problem, we propose the Make Some Noise (MSN) training framework as a replacement for the supervised fine-tuning stage of the large language model. The training method simply introduces some noise at the input for the model to learn the denoising task. It significantly enhances the parallel decoding capability of the model without affecting the original task capability. In addition, we propose a tree-based retrieval-augmented Jacobi (TR-Jacobi) decoding strategy to further improve the inference speed of MSN models. Experiments in both the general and code domains have shown that MSN can improve inference speed by 2.3-2.7x times without compromising model performance. The MSN model also achieves comparable acceleration ratios to the SOTA model with additional model structure on Spec-Bench. | [
"Wang, Yixuan",
"Luo, Xianzhen",
"Wei, Fuxuan",
"Liu, Yijun",
"Zhu, Qingfu",
"Zhang, Xuanyu",
"Yang, Qing",
"Xu, Dongliang",
"Che, Wanxiang"
] | Make Some Noise: Unlocking Language Model Parallel Inference Capability through Noisy Training | emnlp-main.718 | Poster | 2406.17404 | [
""
] | https://huggingface.co/papers/2406.17404 | 0 | 1 | 1 | 9 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.719.bib | https://aclanthology.org/2024.emnlp-main.719/ | @inproceedings{chang-etal-2024-target,
title = "Target-Aware Language Modeling via Granular Data Sampling",
author = "Chang, Ernie and
Lin, Pin-Jie and
Li, Yang and
Zhao, Changsheng and
Kim, Daeil and
Rabatin, Rastislav and
Liu, Zechun and
Shi, Yangyang and
Chandra, Vikas",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.719",
pages = "12927--12935",
abstract = "Language model pretraining generally targets a broad range of use cases and incorporates data from diverse sources. However, there are instances where we desire a model that excels in specific areas without markedly compromising performance in other areas. A cost-effective and straightforward approach is sampling with low-dimensional data features, which allows selecting large-scale pretraining data for domain-specific use cases. In this work, we revisit importance sampling with n-gram features consisting of multi-granular tokens, which strikes a good balance between sentence compression and representation capabilities. We observed the sampled data to have a high correlation with the target downstream task performance *while preserving its effectiveness on other tasks*. This leads to the proposed data sampling paradigm where language models can be pretrained more efficiently on selected documents. On eight benchmarks we demonstrate with {\textasciitilde}1{\%} of the data, pretrained models perform on par with the full RefinedWeb data and outperform randomly selected samples for model sizes ranging from 125M to 1.5B.",
}
| Language model pretraining generally targets a broad range of use cases and incorporates data from diverse sources. However, there are instances where we desire a model that excels in specific areas without markedly compromising performance in other areas. A cost-effective and straightforward approach is sampling with low-dimensional data features, which allows selecting large-scale pretraining data for domain-specific use cases. In this work, we revisit importance sampling with n-gram features consisting of multi-granular tokens, which strikes a good balance between sentence compression and representation capabilities. We observed the sampled data to have a high correlation with the target downstream task performance *while preserving its effectiveness on other tasks*. This leads to the proposed data sampling paradigm where language models can be pretrained more efficiently on selected documents. On eight benchmarks we demonstrate with {\textasciitilde}1{\%} of the data, pretrained models perform on par with the full RefinedWeb data and outperform randomly selected samples for model sizes ranging from 125M to 1.5B. | [
"Chang, Ernie",
"Lin, Pin-Jie",
"Li, Yang",
"Zhao, Changsheng",
"Kim, Daeil",
"Rabatin, Rastislav",
"Liu, Zechun",
"Shi, Yangyang",
"Ch",
"ra, Vikas"
] | Target-Aware Language Modeling via Granular Data Sampling | emnlp-main.719 | Poster | 2409.14705 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.720.bib | https://aclanthology.org/2024.emnlp-main.720/ | @inproceedings{parekh-etal-2024-speed,
title = "{SPEED}++: A Multilingual Event Extraction Framework for Epidemic Prediction and Preparedness",
author = "Parekh, Tanmay and
Kwan, Jeffrey and
Yu, Jiarui and
Johri, Sparsh and
Ahn, Hyosang and
Muppalla, Sreya and
Chang, Kai-Wei and
Wang, Wei and
Peng, Nanyun",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.720",
pages = "12936--12965",
abstract = "Social media is often the first place where communities discuss the latest societal trends. Prior works have utilized this platform to extract epidemic-related information (e.g. infections, preventive measures) to provide early warnings for epidemic prediction. However, these works only focused on English posts, while epidemics can occur anywhere in the world, and early discussions are often in the local, non-English languages. In this work, we introduce the first multilingual Event Extraction (EE) framework SPEED++ for extracting epidemic event information for any disease and language. To this end, we extend a previous epidemic ontology with 20 argument roles; and curate our multilingual EE dataset SPEED++ comprising 5.1K tweets in four languages for four diseases. Annotating data in every language is infeasible; thus we develop zero-shot cross-lingual cross-disease models (i.e., training only on English COVID data) utilizing multilingual pre-training and show their efficacy in extracting epidemic-related events for 65 diverse languages across different diseases. Experiments demonstrate that our framework can provide epidemic warnings for COVID-19 in its earliest stages in Dec 2019 (3 weeks before global discussions) from Chinese Weibo posts without any training in Chinese. Furthermore, we exploit our framework{'}s argument extraction capabilities to aggregate community epidemic discussions like symptoms and cure measures, aiding misinformation detection and public attention monitoring. Overall, we lay a strong foundation for multilingual epidemic preparedness.",
}
| Social media is often the first place where communities discuss the latest societal trends. Prior works have utilized this platform to extract epidemic-related information (e.g. infections, preventive measures) to provide early warnings for epidemic prediction. However, these works only focused on English posts, while epidemics can occur anywhere in the world, and early discussions are often in the local, non-English languages. In this work, we introduce the first multilingual Event Extraction (EE) framework SPEED++ for extracting epidemic event information for any disease and language. To this end, we extend a previous epidemic ontology with 20 argument roles; and curate our multilingual EE dataset SPEED++ comprising 5.1K tweets in four languages for four diseases. Annotating data in every language is infeasible; thus we develop zero-shot cross-lingual cross-disease models (i.e., training only on English COVID data) utilizing multilingual pre-training and show their efficacy in extracting epidemic-related events for 65 diverse languages across different diseases. Experiments demonstrate that our framework can provide epidemic warnings for COVID-19 in its earliest stages in Dec 2019 (3 weeks before global discussions) from Chinese Weibo posts without any training in Chinese. Furthermore, we exploit our framework{'}s argument extraction capabilities to aggregate community epidemic discussions like symptoms and cure measures, aiding misinformation detection and public attention monitoring. Overall, we lay a strong foundation for multilingual epidemic preparedness. | [
"Parekh, Tanmay",
"Kwan, Jeffrey",
"Yu, Jiarui",
"Johri, Sparsh",
"Ahn, Hyosang",
"Muppalla, Sreya",
"Chang, Kai-Wei",
"Wang, Wei",
"Peng, Nanyun"
] | SPEED++: A Multilingual Event Extraction Framework for Epidemic Prediction and Preparedness | emnlp-main.720 | Poster | 2410.18393 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.721.bib | https://aclanthology.org/2024.emnlp-main.721/ | @inproceedings{gul-artzi-2024-cogen,
title = "{C}o{G}en: Learning from Feedback with Coupled Comprehension and Generation",
author = "Gul, Mustafa Omer and
Artzi, Yoav",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.721",
pages = "12966--12982",
abstract = "Systems with both language comprehension and generation capabilities can benefit from the tight connection between the two. This work studies coupling comprehension and generation with focus on continually learning from interaction with users. We propose techniques to tightly integrate the two capabilities for both learning and inference. We situate our studies in two-player reference games, and deploy various models for thousands of interactions with human users, while learning from interaction feedback signals. We show dramatic improvements in performance over time, with comprehension-generation coupling leading to performance improvements up to 26{\%} in absolute terms and up to 17{\%} higher accuracies compared to a non-coupled system. Our analysis also shows coupling has substantial qualitative impact on the system{'}s language, making it significantly more human-like.",
}
| Systems with both language comprehension and generation capabilities can benefit from the tight connection between the two. This work studies coupling comprehension and generation with focus on continually learning from interaction with users. We propose techniques to tightly integrate the two capabilities for both learning and inference. We situate our studies in two-player reference games, and deploy various models for thousands of interactions with human users, while learning from interaction feedback signals. We show dramatic improvements in performance over time, with comprehension-generation coupling leading to performance improvements up to 26{\%} in absolute terms and up to 17{\%} higher accuracies compared to a non-coupled system. Our analysis also shows coupling has substantial qualitative impact on the system{'}s language, making it significantly more human-like. | [
"Gul, Mustafa Omer",
"Artzi, Yoav"
] | CoGen: Learning from Feedback with Coupled Comprehension and Generation | emnlp-main.721 | Poster | 2408.15992 | [
"https://github.com/lil-lab/cogen"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.722.bib | https://aclanthology.org/2024.emnlp-main.722/ | @inproceedings{xiong-etal-2024-unicorn,
title = "{UNICORN}: A Unified Causal Video-Oriented Language-Modeling Framework for Temporal Video-Language Tasks",
author = "Xiong, Yuanhao and
Nie, Yixin and
Liu, Haotian and
Wang, Boxin and
Chen, Jun and
Jin, Rong and
Hsieh, Cho-Jui and
Torresani, Lorenzo and
Lei, Jie",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.722",
pages = "12983--12997",
abstract = "The great success of large language models has encouraged the development of large multimodal models, with a focus on image-language interaction. Despite promising results in various image-language downstream tasks, it is still challenging and unclear how to extend the capabilities of these models to the more complex video domain, especially when dealing with explicit temporal signals. To address the problem in existing large multimodal models, in this paper we adopt visual instruction tuning to build a unified causal video-oriented language modeling framework, named UNICORN. Specifically, we collect a comprehensive dataset under the instruction-following format, and instruction-tune the model accordingly. Experimental results demonstrate that without customized training objectives and intensive pre-training, UNICORN can achieve comparable or better performance on established temporal video-language tasks including moment retrieval, video paragraph captioning and dense video captioning. Moreover, the instruction-tuned model can be used to automatically annotate internet videos with temporally-aligned captions. Compared to commonly used ASR captions, we show that training on our generated captions improves the performance of video-language models on both zero-shot and fine-tuning settings. Source code can be found at https://github.com/xyh97/UNICORN.",
}
| The great success of large language models has encouraged the development of large multimodal models, with a focus on image-language interaction. Despite promising results in various image-language downstream tasks, it is still challenging and unclear how to extend the capabilities of these models to the more complex video domain, especially when dealing with explicit temporal signals. To address the problem in existing large multimodal models, in this paper we adopt visual instruction tuning to build a unified causal video-oriented language modeling framework, named UNICORN. Specifically, we collect a comprehensive dataset under the instruction-following format, and instruction-tune the model accordingly. Experimental results demonstrate that without customized training objectives and intensive pre-training, UNICORN can achieve comparable or better performance on established temporal video-language tasks including moment retrieval, video paragraph captioning and dense video captioning. Moreover, the instruction-tuned model can be used to automatically annotate internet videos with temporally-aligned captions. Compared to commonly used ASR captions, we show that training on our generated captions improves the performance of video-language models on both zero-shot and fine-tuning settings. Source code can be found at https://github.com/xyh97/UNICORN. | [
"Xiong, Yuanhao",
"Nie, Yixin",
"Liu, Haotian",
"Wang, Boxin",
"Chen, Jun",
"Jin, Rong",
"Hsieh, Cho-Jui",
"Torresani, Lorenzo",
"Lei, Jie"
] | UNICORN: A Unified Causal Video-Oriented Language-Modeling Framework for Temporal Video-Language Tasks | emnlp-main.722 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.723.bib | https://aclanthology.org/2024.emnlp-main.723/ | @inproceedings{hobson-etal-2024-story,
title = "Story Morals: Surfacing value-driven narrative schemas using large language models",
author = "Hobson, David G and
Zhou, Haiqi and
Ruths, Derek and
Piper, Andrew",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.723",
pages = "12998--13032",
abstract = "Stories are not only designed to entertain but encode lessons reflecting their authors{'} beliefs about the world. In this paper, we propose a new task of narrative schema labelling based on the concept of {``}story morals{''} to identify the values and lessons conveyed in stories. Using large language models (LLMs) such as GPT-4, we develop methods to automatically extract and validate story morals across a diverse set of narrative genres, including folktales, novels, movies and TV, personal stories from social media and the news. Our approach involves a multi-step prompting sequence to derive morals and validate them through both automated metrics and human assessments. The findings suggest that LLMs can effectively approximate human story moral interpretations and offer a new avenue for computational narrative understanding. By clustering the extracted morals on a sample dataset of folktales from around the world, we highlight the commonalities and distinctiveness of narrative values, providing preliminary insights into the distribution of values across cultures. This work opens up new possibilities for studying narrative schemas and their role in shaping human beliefs and behaviors.",
}
| Stories are not only designed to entertain but encode lessons reflecting their authors{'} beliefs about the world. In this paper, we propose a new task of narrative schema labelling based on the concept of {``}story morals{''} to identify the values and lessons conveyed in stories. Using large language models (LLMs) such as GPT-4, we develop methods to automatically extract and validate story morals across a diverse set of narrative genres, including folktales, novels, movies and TV, personal stories from social media and the news. Our approach involves a multi-step prompting sequence to derive morals and validate them through both automated metrics and human assessments. The findings suggest that LLMs can effectively approximate human story moral interpretations and offer a new avenue for computational narrative understanding. By clustering the extracted morals on a sample dataset of folktales from around the world, we highlight the commonalities and distinctiveness of narrative values, providing preliminary insights into the distribution of values across cultures. This work opens up new possibilities for studying narrative schemas and their role in shaping human beliefs and behaviors. | [
"Hobson, David G",
"Zhou, Haiqi",
"Ruths, Derek",
"Piper, Andrew"
] | Story Morals: Surfacing value-driven narrative schemas using large language models | emnlp-main.723 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.724.bib | https://aclanthology.org/2024.emnlp-main.724/ | @inproceedings{ranjit-etal-2024-oath,
title = "{OATH}-Frames: Characterizing Online Attitudes Towards Homelessness with {LLM} Assistants",
author = "Ranjit, Jaspreet and
Joshi, Brihi and
Dorn, Rebecca and
Petry, Laura and
Koumoundouros, Olga and
Bottarini, Jayne and
Liu, Peichen and
Rice, Eric and
Swayamdipta, Swabha",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.724",
pages = "13033--13059",
abstract = "Warning: Contents of this paper may be upsetting.Public attitudes towards key societal issues, expressed on online media, are of immense value in policy and reform efforts, yet challenging to understand at scale. We study one such social issue: homelessness in the U.S., by leveraging the remarkable capabilities of large language models to assist social work experts in analyzing millions of posts from Twitter. We introduce a framing typology: Online Attitudes Towards Homelessness (OATH) Frames: nine hierarchical frames capturing critiques, responses and perceptions. We release annotations with varying degrees of assistance from language models, with immense benefits in scaling: 6.5{\mbox{$\times$}} speedup in annotation time while only incurring a 3 point F1 reduction in performance with respect to the domain experts. Our experiments demonstrate the value of modeling OATH-Frames over existing sentiment and toxicity classifiers. Our large-scale analysis with predicted OATH-Frames on 2.4M posts on homelessness reveal key trends in attitudes across states, time periods and vulnerable populations, enabling new insights on the issue. Our work provides a general framework to understand nuanced public attitudes at scale, on issues beyond homelessness.",
}
| Warning: Contents of this paper may be upsetting.Public attitudes towards key societal issues, expressed on online media, are of immense value in policy and reform efforts, yet challenging to understand at scale. We study one such social issue: homelessness in the U.S., by leveraging the remarkable capabilities of large language models to assist social work experts in analyzing millions of posts from Twitter. We introduce a framing typology: Online Attitudes Towards Homelessness (OATH) Frames: nine hierarchical frames capturing critiques, responses and perceptions. We release annotations with varying degrees of assistance from language models, with immense benefits in scaling: 6.5{\mbox{$\times$}} speedup in annotation time while only incurring a 3 point F1 reduction in performance with respect to the domain experts. Our experiments demonstrate the value of modeling OATH-Frames over existing sentiment and toxicity classifiers. Our large-scale analysis with predicted OATH-Frames on 2.4M posts on homelessness reveal key trends in attitudes across states, time periods and vulnerable populations, enabling new insights on the issue. Our work provides a general framework to understand nuanced public attitudes at scale, on issues beyond homelessness. | [
"Ranjit, Jaspreet",
"Joshi, Brihi",
"Dorn, Rebecca",
"Petry, Laura",
"Koumoundouros, Olga",
"Bottarini, Jayne",
"Liu, Peichen",
"Rice, Eric",
"Swayamdipta, Swabha"
] | OATH-Frames: Characterizing Online Attitudes Towards Homelessness with LLM Assistants | emnlp-main.724 | Oral | 2406.14883 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.725.bib | https://aclanthology.org/2024.emnlp-main.725/ | @inproceedings{ye-etal-2024-analobench,
title = "{A}nalo{B}ench: Benchmarking the Identification of Abstract and Long-context Analogies",
author = "Ye, Xiao and
Wang, Andrew and
Choi, Jacob and
Lu, Yining and
Sharma, Shreya and
Shen, Lingfeng and
Tiyyala, Vijay Murari and
Andrews, Nicholas and
Khashabi, Daniel",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.725",
pages = "13060--13082",
abstract = "Humans regularly engage in analogical thinking, relating personal experiences to current situations (X is analogous to Y because of Z). Analogical thinking allows humans to solve problems in creative ways, grasp difficult concepts, and articulate ideas more effectively. Can language models (LMs) do the same? To answer this question, we propose AnaloBench, a benchmark to determine analogical reasoning ability in LMs. Our benchmarking approach focuses on aspects of this ability that are common among humans: (i) recalling related experiences from a large amount of information, and (ii) applying analogical reasoning to complex and lengthy scenarios. We collect a set of 340 high quality, human written analogies for use in our benchmark, which constitutes the largest such collection to date. We then test a broad collection of models consisting of 12 open source and 3 proprietary in various sizes and architectures. As in prior results, scaling up LMs results in some performance boosts. Surprisingly, scale offers minimal gains when, (i) analogies involve lengthy scenarios, or (ii) recalling relevant scenarios from a large pool of information, a process analogous to finding a needle in a haystack. We hope these observations encourage further research in this field.",
}
| Humans regularly engage in analogical thinking, relating personal experiences to current situations (X is analogous to Y because of Z). Analogical thinking allows humans to solve problems in creative ways, grasp difficult concepts, and articulate ideas more effectively. Can language models (LMs) do the same? To answer this question, we propose AnaloBench, a benchmark to determine analogical reasoning ability in LMs. Our benchmarking approach focuses on aspects of this ability that are common among humans: (i) recalling related experiences from a large amount of information, and (ii) applying analogical reasoning to complex and lengthy scenarios. We collect a set of 340 high quality, human written analogies for use in our benchmark, which constitutes the largest such collection to date. We then test a broad collection of models consisting of 12 open source and 3 proprietary in various sizes and architectures. As in prior results, scaling up LMs results in some performance boosts. Surprisingly, scale offers minimal gains when, (i) analogies involve lengthy scenarios, or (ii) recalling relevant scenarios from a large pool of information, a process analogous to finding a needle in a haystack. We hope these observations encourage further research in this field. | [
"Ye, Xiao",
"Wang, Andrew",
"Choi, Jacob",
"Lu, Yining",
"Sharma, Shreya",
"Shen, Lingfeng",
"Tiyyala, Vijay Murari",
"Andrews, Nicholas",
"Khashabi, Daniel"
] | AnaloBench: Benchmarking the Identification of Abstract and Long-context Analogies | emnlp-main.725 | Poster | 2402.12370 | [
"https://github.com/jhu-clsp/analogical-reasoning"
] | https://huggingface.co/papers/2402.12370 | 1 | 1 | 0 | 9 | [] | [
"jhu-clsp/AnaloBench"
] | [] | [] | [
"jhu-clsp/AnaloBench"
] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.726.bib | https://aclanthology.org/2024.emnlp-main.726/ | @inproceedings{zhang-etal-2024-scier,
title = "{S}ci{ER}: An Entity and Relation Extraction Dataset for Datasets, Methods, and Tasks in Scientific Documents",
author = "Zhang, Qi and
Chen, Zhijia and
Pan, Huitong and
Caragea, Cornelia and
Latecki, Longin Jan and
Dragut, Eduard",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.726",
pages = "13083--13100",
abstract = "Scientific information extraction (SciIE) is critical for converting unstructured knowledge from scholarly articles into structured data (entities and relations). Several datasets have been proposed for training and validating SciIE models. However, due to the high complexity and cost of annotating scientific texts, those datasets restrict their annotations to specific parts of paper, such as abstracts, resulting in the loss of diverse entity mentions and relations in context. In this paper, we release a new entity and relation extraction dataset for entities related to datasets, methods, and tasks in scientific articles. Our dataset contains 106 manually annotated full-text scientific publications with over 24k entities and 12k relations. To capture the intricate use and interactions among entities in full texts, our dataset contains a fine-grained tag set for relations. Additionally, we provide an out-of-distribution test set to offer a more realistic evaluation. We conduct comprehensive experiments, including state-of-the-art supervised models and our proposed LLM-based baselines, and highlight the challenges presented by our dataset, encouraging the development of innovative models to further the field of SciIE.",
}
| Scientific information extraction (SciIE) is critical for converting unstructured knowledge from scholarly articles into structured data (entities and relations). Several datasets have been proposed for training and validating SciIE models. However, due to the high complexity and cost of annotating scientific texts, those datasets restrict their annotations to specific parts of paper, such as abstracts, resulting in the loss of diverse entity mentions and relations in context. In this paper, we release a new entity and relation extraction dataset for entities related to datasets, methods, and tasks in scientific articles. Our dataset contains 106 manually annotated full-text scientific publications with over 24k entities and 12k relations. To capture the intricate use and interactions among entities in full texts, our dataset contains a fine-grained tag set for relations. Additionally, we provide an out-of-distribution test set to offer a more realistic evaluation. We conduct comprehensive experiments, including state-of-the-art supervised models and our proposed LLM-based baselines, and highlight the challenges presented by our dataset, encouraging the development of innovative models to further the field of SciIE. | [
"Zhang, Qi",
"Chen, Zhijia",
"Pan, Huitong",
"Caragea, Cornelia",
"Latecki, Longin Jan",
"Dragut, Eduard"
] | SciER: An Entity and Relation Extraction Dataset for Datasets, Methods, and Tasks in Scientific Documents | emnlp-main.726 | Poster | 2410.21155 | [
"https://github.com/edzq/SciER"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.727.bib | https://aclanthology.org/2024.emnlp-main.727/ | @inproceedings{godbole-etal-2024-analysis,
title = "Analysis of Plan-based Retrieval for Grounded Text Generation",
author = "Godbole, Ameya and
Monath, Nicholas and
Kim, Seungyeon and
Rawat, Ankit Singh and
McCallum, Andrew and
Zaheer, Manzil",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.727",
pages = "13101--13119",
abstract = "In text generation, hallucinations refer to the generation of seemingly coherent text that contradicts established knowledge. One compelling hypothesis is that hallucinations occur when a language model is given a generation task outside its parametric knowledge (due to rarity, recency, domain, etc.). A common strategy to address this limitation is to infuse the language models with retrieval mechanisms, providing the model with relevant knowledge for the task. In this paper, we leverage the planning capabilities of instruction-tuned LLMs and analyze how planning can be used to guide retrieval to further reduce the frequency of hallucinations. We empirically evaluate several variations of our proposed approach on long-form text generation tasks. By improving the coverage of relevant facts, plan-guided retrieval and generation can produce more informative responses while providing a higher rate of attribution to source documents.",
}
| In text generation, hallucinations refer to the generation of seemingly coherent text that contradicts established knowledge. One compelling hypothesis is that hallucinations occur when a language model is given a generation task outside its parametric knowledge (due to rarity, recency, domain, etc.). A common strategy to address this limitation is to infuse the language models with retrieval mechanisms, providing the model with relevant knowledge for the task. In this paper, we leverage the planning capabilities of instruction-tuned LLMs and analyze how planning can be used to guide retrieval to further reduce the frequency of hallucinations. We empirically evaluate several variations of our proposed approach on long-form text generation tasks. By improving the coverage of relevant facts, plan-guided retrieval and generation can produce more informative responses while providing a higher rate of attribution to source documents. | [
"Godbole, Ameya",
"Monath, Nicholas",
"Kim, Seungyeon",
"Rawat, Ankit Singh",
"McCallum, Andrew",
"Zaheer, Manzil"
] | Analysis of Plan-based Retrieval for Grounded Text Generation | emnlp-main.727 | Poster | 2408.10490 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.728.bib | https://aclanthology.org/2024.emnlp-main.728/ | @inproceedings{chandler-etal-2024-detecting,
title = "Detecting Errors through Ensembling Prompts ({DEEP}): An End-to-End {LLM} Framework for Detecting Factual Errors",
author = "Chandler, Alex and
Surve, Devesh and
Su, Hui",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.728",
pages = "13120--13133",
abstract = "Accurate text summarization is one of the most common and important tasks performed by Large Language Models, where the costs of human review for an entire document may be high, but the costs of errors in summarization may be even greater. We propose Detecting Errors through Ensembling Prompts (DEEP) - an end-to-end large language model framework for detecting factual errors in text summarization. Our framework uses a diverse set of LLM prompts to identify factual inconsistencies, treating their outputs as binary features, which are then fed into ensembling models. We then calibrate the ensembled models to produce empirically accurate probabilities that a text is factually consistent or free of hallucination. We demonstrate that prior models for detecting factual errors in summaries perform significantly worse without optimizing the thresholds on subsets of the evaluated dataset. Our framework achieves state-of-the-art (SOTA) balanced accuracy on the AggreFact-XSUM FTSOTA, TofuEval Summary-Level, and HaluEval Summarization benchmarks in detecting factual errors within transformer-generated text summaries. It does so without any fine-tuning of the language model or reliance on thresholding techniques not available in practical settings.",
}
| Accurate text summarization is one of the most common and important tasks performed by Large Language Models, where the costs of human review for an entire document may be high, but the costs of errors in summarization may be even greater. We propose Detecting Errors through Ensembling Prompts (DEEP) - an end-to-end large language model framework for detecting factual errors in text summarization. Our framework uses a diverse set of LLM prompts to identify factual inconsistencies, treating their outputs as binary features, which are then fed into ensembling models. We then calibrate the ensembled models to produce empirically accurate probabilities that a text is factually consistent or free of hallucination. We demonstrate that prior models for detecting factual errors in summaries perform significantly worse without optimizing the thresholds on subsets of the evaluated dataset. Our framework achieves state-of-the-art (SOTA) balanced accuracy on the AggreFact-XSUM FTSOTA, TofuEval Summary-Level, and HaluEval Summarization benchmarks in detecting factual errors within transformer-generated text summaries. It does so without any fine-tuning of the language model or reliance on thresholding techniques not available in practical settings. | [
"Ch",
"ler, Alex",
"Surve, Devesh",
"Su, Hui"
] | Detecting Errors through Ensembling Prompts (DEEP): An End-to-End LLM Framework for Detecting Factual Errors | emnlp-main.728 | Poster | 2406.13009 | [
"https://github.com/achandlr/DEEP"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.729.bib | https://aclanthology.org/2024.emnlp-main.729/ | @inproceedings{dang-etal-2024-rlhf,
title = "{RLHF} Can Speak Many Languages: Unlocking Multilingual Preference Optimization for {LLM}s",
author = {Dang, John and
Ahmadian, Arash and
Marchisio, Kelly and
Kreutzer, Julia and
{\"U}st{\"u}n, Ahmet and
Hooker, Sara},
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.729",
pages = "13134--13156",
abstract = "Preference optimization techniques have become a standard final stage for training state-of-art large language models (LLMs). However, despite widespread adoption, the vast majority of work to-date has focused on a small set of high-resource languages like English and Chinese. This captures a small fraction of the languages in the world, but also makes it unclear which aspects of current state-of-the-art research transfer to a multilingual setting. In this work, we perform an exhaustive study to achieve a new state of the art in aligning multilingual LLMs. We introduce a novel, scalable method for generating high-quality multilingual feedback data to balance data coverage. We establish the benefits of cross-lingual transfer and increased dataset size in preference training. Our preference-trained model achieves a 54.4{\%} win-rate against Aya 23 8B, the current state-of-the-art multilingual LLM in its parameter class, and a 69.5{\%} win-rate or higher against widely used models like Gemma, Mistral and Llama 3. As a result of our efforts, we expand the frontier of alignment techniques to 23 languages, covering approximately half of the world{'}s population.",
}
| Preference optimization techniques have become a standard final stage for training state-of-art large language models (LLMs). However, despite widespread adoption, the vast majority of work to-date has focused on a small set of high-resource languages like English and Chinese. This captures a small fraction of the languages in the world, but also makes it unclear which aspects of current state-of-the-art research transfer to a multilingual setting. In this work, we perform an exhaustive study to achieve a new state of the art in aligning multilingual LLMs. We introduce a novel, scalable method for generating high-quality multilingual feedback data to balance data coverage. We establish the benefits of cross-lingual transfer and increased dataset size in preference training. Our preference-trained model achieves a 54.4{\%} win-rate against Aya 23 8B, the current state-of-the-art multilingual LLM in its parameter class, and a 69.5{\%} win-rate or higher against widely used models like Gemma, Mistral and Llama 3. As a result of our efforts, we expand the frontier of alignment techniques to 23 languages, covering approximately half of the world{'}s population. | [
"Dang, John",
"Ahmadian, Arash",
"Marchisio, Kelly",
"Kreutzer, Julia",
"{\\\"U}st{\\\"u}n, Ahmet",
"Hooker, Sara"
] | RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs | emnlp-main.729 | Poster | 2407.02552 | [
""
] | https://huggingface.co/papers/2407.02552 | 5 | 4 | 0 | 6 | [
"CohereForAI/aya-expanse-8b",
"CohereForAI/aya-expanse-32b",
"QuantFactory/aya-expanse-8b-GGUF",
"lucyknada/CohereForAI_aya-expanse-8b-exl2",
"lucyknada/CohereForAI_aya-expanse-32b-exl2",
"adamo1139/aya-expanse-8b-ungated",
"jth01/aya-expanse-8b-5.0bpw-exl2",
"duyntnet/aya-expanse-8b-imatrix-GGUF",
"Andrewwwwww/aya-expanse-32b",
"Svngoku/Aya-Expanse-8B-French",
"duyntnet/aya-expanse-32b-imatrix-GGUF",
"Jellon/aya-expanse-32b-exl2-4bpw",
"Jellon/aya-expanse-32b-exl2-6bpw",
"adamo1139/aya-expanse-32b-ungated",
"Svngoku/French-Aya-Expanse-8B",
"RichardErkhov/adamo1139_-_aya-expanse-32b-ungated-gguf"
] | [] | [
"CohereForAI/aya_expanse",
"logikon/open_cot_leaderboard",
"cot-leaderboard/open-cot-dashboard",
"Rijgersberg/Aya-Expanse-8B",
"IllyrianSpace/aya_expanse",
"Svngoku/Aya-Expanse-8B",
"Anupam251272/AJ-Chat",
"arnavnextai/Aya-Expanse-8B"
] | [
"CohereForAI/aya-expanse-8b",
"CohereForAI/aya-expanse-32b",
"QuantFactory/aya-expanse-8b-GGUF",
"lucyknada/CohereForAI_aya-expanse-8b-exl2",
"lucyknada/CohereForAI_aya-expanse-32b-exl2",
"adamo1139/aya-expanse-8b-ungated",
"jth01/aya-expanse-8b-5.0bpw-exl2",
"duyntnet/aya-expanse-8b-imatrix-GGUF",
"Andrewwwwww/aya-expanse-32b",
"Svngoku/Aya-Expanse-8B-French",
"duyntnet/aya-expanse-32b-imatrix-GGUF",
"Jellon/aya-expanse-32b-exl2-4bpw",
"Jellon/aya-expanse-32b-exl2-6bpw",
"adamo1139/aya-expanse-32b-ungated",
"Svngoku/French-Aya-Expanse-8B",
"RichardErkhov/adamo1139_-_aya-expanse-32b-ungated-gguf"
] | [] | [
"CohereForAI/aya_expanse",
"logikon/open_cot_leaderboard",
"cot-leaderboard/open-cot-dashboard",
"Rijgersberg/Aya-Expanse-8B",
"IllyrianSpace/aya_expanse",
"Svngoku/Aya-Expanse-8B",
"Anupam251272/AJ-Chat",
"arnavnextai/Aya-Expanse-8B"
] | 1 |
https://aclanthology.org/2024.emnlp-main.730.bib | https://aclanthology.org/2024.emnlp-main.730/ | @inproceedings{lei-huang-2024-boosting,
title = "Boosting Logical Fallacy Reasoning in {LLM}s via Logical Structure Tree",
author = "Lei, Yuanyuan and
Huang, Ruihong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.730",
pages = "13157--13173",
abstract = "Logical fallacy uses invalid or faulty reasoning in the construction of a statement. Despite the prevalence and harmfulness of logical fallacies, detecting and classifying logical fallacies still remains a challenging task. We observe that logical fallacies often use connective words to indicate an intended logical relation between two arguments, while the argument semantics does not actually support the logical relation. Inspired by this observation, we propose to build a logical structure tree to explicitly represent and track the hierarchical logic flow among relation connectives and their arguments in a statement. Specifically, this logical structure tree is constructed in an unsupervised manner guided by the constituency tree and a taxonomy of connectives for ten common logical relations, with relation connectives as non-terminal nodes and textual arguments as terminal nodes, and the latter are mostly elementary discourse units. We further develop two strategies to incorporate the logical structure tree into LLMs for fallacy reasoning. Firstly, we transform the tree into natural language descriptions and feed the textualized tree into LLMs as a part of the hard text prompt. Secondly, we derive a relation-aware tree embedding and insert the tree embedding into LLMs as a soft prompt. Experiments on benchmark datasets demonstrate that our approach based on logical structure tree significantly improves precision and recall for both fallacy detection and fallacy classification.",
}
| Logical fallacy uses invalid or faulty reasoning in the construction of a statement. Despite the prevalence and harmfulness of logical fallacies, detecting and classifying logical fallacies still remains a challenging task. We observe that logical fallacies often use connective words to indicate an intended logical relation between two arguments, while the argument semantics does not actually support the logical relation. Inspired by this observation, we propose to build a logical structure tree to explicitly represent and track the hierarchical logic flow among relation connectives and their arguments in a statement. Specifically, this logical structure tree is constructed in an unsupervised manner guided by the constituency tree and a taxonomy of connectives for ten common logical relations, with relation connectives as non-terminal nodes and textual arguments as terminal nodes, and the latter are mostly elementary discourse units. We further develop two strategies to incorporate the logical structure tree into LLMs for fallacy reasoning. Firstly, we transform the tree into natural language descriptions and feed the textualized tree into LLMs as a part of the hard text prompt. Secondly, we derive a relation-aware tree embedding and insert the tree embedding into LLMs as a soft prompt. Experiments on benchmark datasets demonstrate that our approach based on logical structure tree significantly improves precision and recall for both fallacy detection and fallacy classification. | [
"Lei, Yuanyuan",
"Huang, Ruihong"
] | Boosting Logical Fallacy Reasoning in LLMs via Logical Structure Tree | emnlp-main.730 | Poster | 2410.12048 | [
"https://github.com/yuanyuanlei-nlp/logical_fallacy_emnlp_2024"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.731.bib | https://aclanthology.org/2024.emnlp-main.731/ | @inproceedings{fagnou-etal-2024-chain,
title = "Chain and Causal Attention for Efficient Entity Tracking",
author = "Fagnou, Erwan and
Caillon, Paul and
Delattre, Blaise and
Allauzen, Alexandre",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.731",
pages = "13174--13188",
abstract = "This paper investigates the limitations of transformers for entity-tracking tasks in large language models. We identify a theoretical constraint, showing that transformers require at least $\log_2 (n+1)$ layers to handle entity tracking with $n$ state changes. To address this issue, we propose an efficient and frugal enhancement to the standard attention mechanism, enabling it to manage long-term dependencies more efficiently. By considering attention as an adjacency matrix, our model can track entity states with a single layer.Empirical results demonstrate significant improvements in entity tracking datasets while keeping competitive performance on standard natural language modeling. Our modified attention allows us to achieve the same performance with drastically fewer layers. Additionally, our enhanced mechanism reveals structured internal representations of attention. Extensive experiments on both toy and complex datasets validate our approach. Our contributions include theoretical insights, an improved attention mechanism, and empirical validation.",
}
| This paper investigates the limitations of transformers for entity-tracking tasks in large language models. We identify a theoretical constraint, showing that transformers require at least $\log_2 (n+1)$ layers to handle entity tracking with $n$ state changes. To address this issue, we propose an efficient and frugal enhancement to the standard attention mechanism, enabling it to manage long-term dependencies more efficiently. By considering attention as an adjacency matrix, our model can track entity states with a single layer.Empirical results demonstrate significant improvements in entity tracking datasets while keeping competitive performance on standard natural language modeling. Our modified attention allows us to achieve the same performance with drastically fewer layers. Additionally, our enhanced mechanism reveals structured internal representations of attention. Extensive experiments on both toy and complex datasets validate our approach. Our contributions include theoretical insights, an improved attention mechanism, and empirical validation. | [
"Fagnou, Erwan",
"Caillon, Paul",
"Delattre, Blaise",
"Allauzen, Alex",
"re"
] | Chain and Causal Attention for Efficient Entity Tracking | emnlp-main.731 | Poster | 2410.05565 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.732.bib | https://aclanthology.org/2024.emnlp-main.732/ | @inproceedings{zeng-etal-2024-beear,
title = "{BEEAR}: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models",
author = "Zeng, Yi and
Sun, Weiyu and
Huynh, Tran and
Song, Dawn and
Li, Bo and
Jia, Ruoxi",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.732",
pages = "13189--13215",
abstract = "Safety backdoor attacks in large language models (LLMs) enable harmful behaviors to be stealthily triggered while evading detection during normal interactions. The high dimensionality of the trigger search space and the diverse range of potential malicious behaviors in LLMs make this a critical open problem. This paper presents BEEAR, a novel mitigation method based on a key insight: backdoor triggers induce a uniform drift in the model{'}s embedding space, irrespective of the trigger{'}s form or targeted behavior. Leveraging this observation, we introduce a bi-level optimization approach. The inner level identifies universal perturbations to the decoder{'}s embeddings that steer the model towards defender-defined unwanted behaviors; the outer level fine-tunes the model to reinforce safe behaviors against these perturbations. Our experiments demonstrate the effectiveness of this approach, reducing the success rate of safety backdoor attacks from over 95{\%} to {\textless}1{\%} for general harmful behaviors and from 47{\%} to 0{\%} for Sleeper Agents, without compromising the model{'}s helpfulness. Notably, our method relies only on defender-defined sets of safe and unwanted behaviors without any assumptions about the trigger location or attack mechanism. This work represents the first practical framework to counter safety backdoors in LLMs and provides a foundation for future advancements in AI safety and security.",
}
| Safety backdoor attacks in large language models (LLMs) enable harmful behaviors to be stealthily triggered while evading detection during normal interactions. The high dimensionality of the trigger search space and the diverse range of potential malicious behaviors in LLMs make this a critical open problem. This paper presents BEEAR, a novel mitigation method based on a key insight: backdoor triggers induce a uniform drift in the model{'}s embedding space, irrespective of the trigger{'}s form or targeted behavior. Leveraging this observation, we introduce a bi-level optimization approach. The inner level identifies universal perturbations to the decoder{'}s embeddings that steer the model towards defender-defined unwanted behaviors; the outer level fine-tunes the model to reinforce safe behaviors against these perturbations. Our experiments demonstrate the effectiveness of this approach, reducing the success rate of safety backdoor attacks from over 95{\%} to {\textless}1{\%} for general harmful behaviors and from 47{\%} to 0{\%} for Sleeper Agents, without compromising the model{'}s helpfulness. Notably, our method relies only on defender-defined sets of safe and unwanted behaviors without any assumptions about the trigger location or attack mechanism. This work represents the first practical framework to counter safety backdoors in LLMs and provides a foundation for future advancements in AI safety and security. | [
"Zeng, Yi",
"Sun, Weiyu",
"Huynh, Tran",
"Song, Dawn",
"Li, Bo",
"Jia, Ruoxi"
] | BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models | emnlp-main.732 | Poster | 2406.17092 | [
"https://github.com/reds-lab/beear"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.733.bib | https://aclanthology.org/2024.emnlp-main.733/ | @inproceedings{hu-etal-2024-bayesian,
title = "A {B}ayesian Approach to Harnessing the Power of {LLM}s in Authorship Attribution",
author = "Hu, Zhengmian and
Zheng, Tong and
Huang, Heng",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.733",
pages = "13216--13227",
abstract = "Authorship attribution aims to identify the origin or author of a document. Traditional approaches have heavily relied on manual features and fail to capture long-range correlations, limiting their effectiveness. Recent advancements leverage text embeddings from pre-trained language models, which require significant fine-tuning on labeled data, posing challenges in data dependency and limited interpretability. Large Language Models (LLMs), with their deep reasoning capabilities and ability to maintain long-range textual associations, offer a promising alternative. This study explores the potential of pre-trained LLMs in one-shot authorship attribution, specifically utilizing Bayesian approaches and probability outputs of LLMs. Our methodology calculates the probability that a text entails previous writings of an author, reflecting a more nuanced understanding of authorship. By utilizing only pre-trained models such as Llama-3-70B, our results on the IMDb and blog datasets show an impressive 85{\%} accuracy in one-shot authorship classification across ten authors. Our findings set new baselines for one-shot authorship analysis using LLMs and expand the application scope of these models in forensic linguistics. This work also includes extensive ablation studies to validate our approach.",
}
| Authorship attribution aims to identify the origin or author of a document. Traditional approaches have heavily relied on manual features and fail to capture long-range correlations, limiting their effectiveness. Recent advancements leverage text embeddings from pre-trained language models, which require significant fine-tuning on labeled data, posing challenges in data dependency and limited interpretability. Large Language Models (LLMs), with their deep reasoning capabilities and ability to maintain long-range textual associations, offer a promising alternative. This study explores the potential of pre-trained LLMs in one-shot authorship attribution, specifically utilizing Bayesian approaches and probability outputs of LLMs. Our methodology calculates the probability that a text entails previous writings of an author, reflecting a more nuanced understanding of authorship. By utilizing only pre-trained models such as Llama-3-70B, our results on the IMDb and blog datasets show an impressive 85{\%} accuracy in one-shot authorship classification across ten authors. Our findings set new baselines for one-shot authorship analysis using LLMs and expand the application scope of these models in forensic linguistics. This work also includes extensive ablation studies to validate our approach. | [
"Hu, Zhengmian",
"Zheng, Tong",
"Huang, Heng"
] | A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution | emnlp-main.733 | Poster | 2410.21716 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.734.bib | https://aclanthology.org/2024.emnlp-main.734/ | @inproceedings{wang-etal-2024-fac2e,
title = "{FAC}$^2${E}: Better Understanding Large Language Model Capabilities by Dissociating Language and Cognition",
author = "Wang, Xiaoqiang and
Wu, Lingfei and
Ma, Tengfei and
Liu, Bang",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.734",
pages = "13228--13243",
abstract = "Large language models (LLMs) are primarily evaluated by overall performance on various text understanding and generation tasks. However, such a paradigm fails to comprehensively differentiate the fine-grained language and cognitive skills, rendering the lack of sufficient interpretation to LLMs{'} capabilities. In this paper, we present FAC$^2$E, a framework for Fine-grAined and Cognition-grounded LLMs{'} Capability Evaluation. Specifically, we formulate LLMs{'} evaluation in a multi-dimensional and explainable manner by dissociating the language-related capabilities and the cognition-related ones. Besides, through extracting the intermediate reasoning from LLMs, we further break down the process of applying a specific capability into three sub-steps: recalling relevant knowledge, utilizing knowledge, and solving problems. Finally, FAC$^2$E evaluates each sub-step of each fine-grained capability, providing a two-faceted diagnosis for LLMs. Utilizing FAC$^2$E, we identify a common shortfall in knowledge utilization among models and propose a straightforward, knowledge-enhanced method to mitigate this issue. Our results not only showcase promising performance enhancements but also highlight a direction for future LLM advancements.",
}
| Large language models (LLMs) are primarily evaluated by overall performance on various text understanding and generation tasks. However, such a paradigm fails to comprehensively differentiate the fine-grained language and cognitive skills, rendering the lack of sufficient interpretation to LLMs{'} capabilities. In this paper, we present FAC$^2$E, a framework for Fine-grAined and Cognition-grounded LLMs{'} Capability Evaluation. Specifically, we formulate LLMs{'} evaluation in a multi-dimensional and explainable manner by dissociating the language-related capabilities and the cognition-related ones. Besides, through extracting the intermediate reasoning from LLMs, we further break down the process of applying a specific capability into three sub-steps: recalling relevant knowledge, utilizing knowledge, and solving problems. Finally, FAC$^2$E evaluates each sub-step of each fine-grained capability, providing a two-faceted diagnosis for LLMs. Utilizing FAC$^2$E, we identify a common shortfall in knowledge utilization among models and propose a straightforward, knowledge-enhanced method to mitigate this issue. Our results not only showcase promising performance enhancements but also highlight a direction for future LLM advancements. | [
"Wang, Xiaoqiang",
"Wu, Lingfei",
"Ma, Tengfei",
"Liu, Bang"
] | FAC^2E: Better Understanding Large Language Model Capabilities by Dissociating Language and Cognition | emnlp-main.734 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.735.bib | https://aclanthology.org/2024.emnlp-main.735/ | @inproceedings{mahmud-marculescu-2024-opensep,
title = "{O}pen{S}ep: Leveraging Large Language Models with Textual Inversion for Open World Audio Separation",
author = "Mahmud, Tanvir and
Marculescu, Diana",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.735",
pages = "13244--13260",
abstract = "Audio separation in real-world scenarios, where mixtures contain a variable number of sources, presents significant challenges due to limitations of existing models, such as over-separation, under-separation, and dependence on predefined training sources. We propose OpenSep, a novel framework that leverages large language models (LLMs) for automated audio separation, eliminating the need for manual intervention and overcoming source limitations. OpenSep uses textual inversion to generate captions from audio mixtures with off-the-shelf audio captioning models, effectively parsing the sound sources present. It then employs few-shot LLM prompting to extract detailed audio properties of each parsed source, facilitating separation in unseen mixtures. Additionally, we introduce a multi-level extension of the mix-and-separate training framework to enhance modality alignment by separating single source sounds and mixtures simultaneously. Extensive experiments demonstrate OpenSep{'}s superiority in precisely separating new, unseen, and variable sources in challenging mixtures, outperforming SOTA baseline methods. Code is released at https://github.com/tanvir-utexas/OpenSep.git.",
}
| Audio separation in real-world scenarios, where mixtures contain a variable number of sources, presents significant challenges due to limitations of existing models, such as over-separation, under-separation, and dependence on predefined training sources. We propose OpenSep, a novel framework that leverages large language models (LLMs) for automated audio separation, eliminating the need for manual intervention and overcoming source limitations. OpenSep uses textual inversion to generate captions from audio mixtures with off-the-shelf audio captioning models, effectively parsing the sound sources present. It then employs few-shot LLM prompting to extract detailed audio properties of each parsed source, facilitating separation in unseen mixtures. Additionally, we introduce a multi-level extension of the mix-and-separate training framework to enhance modality alignment by separating single source sounds and mixtures simultaneously. Extensive experiments demonstrate OpenSep{'}s superiority in precisely separating new, unseen, and variable sources in challenging mixtures, outperforming SOTA baseline methods. Code is released at https://github.com/tanvir-utexas/OpenSep.git. | [
"Mahmud, Tanvir",
"Marculescu, Diana"
] | OpenSep: Leveraging Large Language Models with Textual Inversion for Open World Audio Separation | emnlp-main.735 | Poster | 2409.19270 | [
"https://github.com/tanvir-utexas/opensep"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.736.bib | https://aclanthology.org/2024.emnlp-main.736/ | @inproceedings{huang-etal-2024-language,
title = "Language Concept Erasure for Language-invariant Dense Retrieval",
author = "Huang, Zhiqi and
Yu, Puxuan and
Ravfogel, Shauli and
Allan, James",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.736",
pages = "13261--13273",
abstract = "Multilingual models aim for language-invariant representations but still prominently encode language identity. This, along with the scarcity of high-quality parallel retrieval data, limits their performance in retrieval. We introduce LANCER, a multi-task learning framework that improves language-invariant dense retrieval by reducing language-specific signals in the embedding space. Leveraging the notion of linear concept erasure, we design a loss function that penalizes cross-correlation between representations and their language labels. LANCER leverages only English retrieval data and general multilingual corpora, training models to focus on language-invariant retrieval by semantic similarity without necessitating a vast parallel corpus. Experimental results on various datasets show our method consistently improves over baselines, with extensive analyses demonstrating greater language agnosticism.",
}
| Multilingual models aim for language-invariant representations but still prominently encode language identity. This, along with the scarcity of high-quality parallel retrieval data, limits their performance in retrieval. We introduce LANCER, a multi-task learning framework that improves language-invariant dense retrieval by reducing language-specific signals in the embedding space. Leveraging the notion of linear concept erasure, we design a loss function that penalizes cross-correlation between representations and their language labels. LANCER leverages only English retrieval data and general multilingual corpora, training models to focus on language-invariant retrieval by semantic similarity without necessitating a vast parallel corpus. Experimental results on various datasets show our method consistently improves over baselines, with extensive analyses demonstrating greater language agnosticism. | [
"Huang, Zhiqi",
"Yu, Puxuan",
"Ravfogel, Shauli",
"Allan, James"
] | Language Concept Erasure for Language-invariant Dense Retrieval | emnlp-main.736 | Oral | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.737.bib | https://aclanthology.org/2024.emnlp-main.737/ | @inproceedings{wang-etal-2024-learning-personalized,
title = "Learning Personalized Alignment for Evaluating Open-ended Text Generation",
author = "Wang, Danqing and
Yang, Kevin and
Zhu, Hanlin and
Yang, Xiaomeng and
Cohen, Andrew and
Li, Lei and
Tian, Yuandong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.737",
pages = "13274--13292",
abstract = "Recent research has increasingly focused on evaluating large language models{'} (LLMs) alignment with diverse human values and preferences, particularly for open-ended tasks like story generation. Traditional evaluation metrics rely heavily on lexical similarity with human-written references, often showing poor correlation with human judgments and failing to account for alignment with the diversity of human preferences. To address these challenges, we introduce PerSE, an interpretable evaluation framework designed to assess alignment with specific human preferences. It is tuned to infer specific preferences from an in-context personal profile and evaluate the alignment between the generated content and personal preferences. PerSE enhances interpretability by providing detailed comments and fine-grained scoring, facilitating more personalized content generation. Our 13B LLaMA-2-based PerSE shows a 15.8{\%} increase in Kendall correlation and a 13.7{\%} rise in accuracy with zero-shot reviewers compared to GPT-4. It also outperforms GPT-4 by 46.01{\%} in Kendall correlation on new domains, indicating its transferability",
}
| Recent research has increasingly focused on evaluating large language models{'} (LLMs) alignment with diverse human values and preferences, particularly for open-ended tasks like story generation. Traditional evaluation metrics rely heavily on lexical similarity with human-written references, often showing poor correlation with human judgments and failing to account for alignment with the diversity of human preferences. To address these challenges, we introduce PerSE, an interpretable evaluation framework designed to assess alignment with specific human preferences. It is tuned to infer specific preferences from an in-context personal profile and evaluate the alignment between the generated content and personal preferences. PerSE enhances interpretability by providing detailed comments and fine-grained scoring, facilitating more personalized content generation. Our 13B LLaMA-2-based PerSE shows a 15.8{\%} increase in Kendall correlation and a 13.7{\%} rise in accuracy with zero-shot reviewers compared to GPT-4. It also outperforms GPT-4 by 46.01{\%} in Kendall correlation on new domains, indicating its transferability | [
"Wang, Danqing",
"Yang, Kevin",
"Zhu, Hanlin",
"Yang, Xiaomeng",
"Cohen, Andrew",
"Li, Lei",
"Tian, Yu",
"ong"
] | Learning Personalized Alignment for Evaluating Open-ended Text Generation | emnlp-main.737 | Poster | 2310.03304 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.738.bib | https://aclanthology.org/2024.emnlp-main.738/ | @inproceedings{zhou-etal-2024-large-language,
title = "Large Language Models Are Involuntary Truth-Tellers: Exploiting Fallacy Failure for Jailbreak Attacks",
author = "Zhou, Yue and
Zou, Henry Peng and
Di Eugenio, Barbara and
Zhang, Yang",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.738",
pages = "13293--13304",
abstract = "We find that language models have difficulties generating fallacious and deceptive reasoning. When asked to generate deceptive outputs, language models tend to leak honest counterparts but believe them to be false. Exploiting this deficiency, we propose a jailbreak attack method that elicits an aligned language model for malicious output. Specifically, we query the model to generate a fallacious yet deceptively real procedure for the harmful behavior. Since a fallacious procedure is generally considered fake and thus harmless by LLMs, it helps bypass the safeguard mechanism. Yet the output is factually harmful since the LLM cannot fabricate fallacious solutions but proposes truthful ones. We evaluate our approach over five safety-aligned large language models, comparing four previous jailbreak methods, and show that our approach achieves competitive performance with more harmful outputs. We believe the findings could be extended beyond model safety, such as self-verification and hallucination.",
}
| We find that language models have difficulties generating fallacious and deceptive reasoning. When asked to generate deceptive outputs, language models tend to leak honest counterparts but believe them to be false. Exploiting this deficiency, we propose a jailbreak attack method that elicits an aligned language model for malicious output. Specifically, we query the model to generate a fallacious yet deceptively real procedure for the harmful behavior. Since a fallacious procedure is generally considered fake and thus harmless by LLMs, it helps bypass the safeguard mechanism. Yet the output is factually harmful since the LLM cannot fabricate fallacious solutions but proposes truthful ones. We evaluate our approach over five safety-aligned large language models, comparing four previous jailbreak methods, and show that our approach achieves competitive performance with more harmful outputs. We believe the findings could be extended beyond model safety, such as self-verification and hallucination. | [
"Zhou, Yue",
"Zou, Henry Peng",
"Di Eugenio, Barbara",
"Zhang, Yang"
] | Large Language Models Are Involuntary Truth-Tellers: Exploiting Fallacy Failure for Jailbreak Attacks | emnlp-main.738 | Poster | 2407.00869 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.739.bib | https://aclanthology.org/2024.emnlp-main.739/ | @inproceedings{zeng-etal-2024-turn,
title = "Turn Waste into Worth: Rectifying Top-$k$ Router of {M}o{E}",
author = "Zeng, Zhiyuan and
Guo, Qipeng and
Fei, Zhaoye and
Yin, Zhangyue and
Zhou, Yunhua and
Li, Linyang and
Sun, Tianxiang and
Yan, Hang and
Lin, Dahua and
Qiu, Xipeng",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.739",
pages = "13305--13320",
abstract = "Sparse Mixture of Experts (MoE) models are popular for training large language models due to their computational efficiency. However, the commonly used top-$k$ routing mechanism suffers from redundancy computation and memory costs due to the unbalanced routing. Some experts are overflow, where the exceeding tokens are dropped. While some experts are empty, which are padded with zeros, negatively impacting model performance. To address the dropped tokens and padding, we propose the Rectify-Router, comprising the Intra-GPU Rectification and the Fill-in Rectification. The Intra-GPU Rectification handles dropped tokens, efficiently routing them to experts within the GPU where they are located to avoid inter-GPU communication. The Fill-in Rectification addresses padding by replacing padding tokens with the tokens that have high routing scores. Our experimental results demonstrate that the Intra-GPU Rectification and the Fill-in Rectification effectively handle dropped tokens and padding, respectively. Furthermore, the combination of them achieves superior performance, surpassing the accuracy of the vanilla top-1 router by 4.7{\%}.",
}
| Sparse Mixture of Experts (MoE) models are popular for training large language models due to their computational efficiency. However, the commonly used top-$k$ routing mechanism suffers from redundancy computation and memory costs due to the unbalanced routing. Some experts are overflow, where the exceeding tokens are dropped. While some experts are empty, which are padded with zeros, negatively impacting model performance. To address the dropped tokens and padding, we propose the Rectify-Router, comprising the Intra-GPU Rectification and the Fill-in Rectification. The Intra-GPU Rectification handles dropped tokens, efficiently routing them to experts within the GPU where they are located to avoid inter-GPU communication. The Fill-in Rectification addresses padding by replacing padding tokens with the tokens that have high routing scores. Our experimental results demonstrate that the Intra-GPU Rectification and the Fill-in Rectification effectively handle dropped tokens and padding, respectively. Furthermore, the combination of them achieves superior performance, surpassing the accuracy of the vanilla top-1 router by 4.7{\%}. | [
"Zeng, Zhiyuan",
"Guo, Qipeng",
"Fei, Zhaoye",
"Yin, Zhangyue",
"Zhou, Yunhua",
"Li, Linyang",
"Sun, Tianxiang",
"Yan, Hang",
"Lin, Dahua",
"Qiu, Xipeng"
] | Turn Waste into Worth: Rectifying Top-k Router of MoE | emnlp-main.739 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.740.bib | https://aclanthology.org/2024.emnlp-main.740/ | @inproceedings{taveekitworachai-etal-2024-null,
title = "Null-Shot Prompting: Rethinking Prompting Large Language Models With Hallucination",
author = "Taveekitworachai, Pittawat and
Abdullah, Febri and
Thawonmas, Ruck",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.740",
pages = "13321--13361",
abstract = "This paper presents a series of investigations into an interesting phenomenon where we observe performance increases in large language models (LLMs) when providing a prompt that causes and exploits hallucination. We propose null-shot prompting, a counter-intuitive approach where we intentionally instruct LLMs to look at and utilize information from a null section. We investigate null-shot prompting on a wide range of tasks, including arithmetic reasoning, commonsense reasoning, and reading comprehension. We observe a substantial increase in performance in arithmetic reasoning tasks for various models, with up to a 44.62{\%} increase compared to a baseline in one model. Therefore, we investigate deeper into this task by utilizing a more challenging mathematics problem-solving benchmark. We observe that LLMs benefit from hallucination in null-shot prompting in this task and discuss the mathematical topics that benefit the most from introducing hallucination in the prompt. We continue our investigation by evaluating hallucination detection abilities of the LLMs when using null-shot prompting. We find surprising results where hallucination in prompts can improve hallucination detection abilities of many LLMs. We also examine the effects of introducing both reasoning, which is known to mitigate hallucination, and hallucination simultaneously in the prompt and observe another surprising turn for the mathematics problem-solving benchmark with many performance improvements. We hope this paper will spark more interest, investigations, and discussions on how hallucination in prompts LLMs and even bolsters them in certain cases.",
}
| This paper presents a series of investigations into an interesting phenomenon where we observe performance increases in large language models (LLMs) when providing a prompt that causes and exploits hallucination. We propose null-shot prompting, a counter-intuitive approach where we intentionally instruct LLMs to look at and utilize information from a null section. We investigate null-shot prompting on a wide range of tasks, including arithmetic reasoning, commonsense reasoning, and reading comprehension. We observe a substantial increase in performance in arithmetic reasoning tasks for various models, with up to a 44.62{\%} increase compared to a baseline in one model. Therefore, we investigate deeper into this task by utilizing a more challenging mathematics problem-solving benchmark. We observe that LLMs benefit from hallucination in null-shot prompting in this task and discuss the mathematical topics that benefit the most from introducing hallucination in the prompt. We continue our investigation by evaluating hallucination detection abilities of the LLMs when using null-shot prompting. We find surprising results where hallucination in prompts can improve hallucination detection abilities of many LLMs. We also examine the effects of introducing both reasoning, which is known to mitigate hallucination, and hallucination simultaneously in the prompt and observe another surprising turn for the mathematics problem-solving benchmark with many performance improvements. We hope this paper will spark more interest, investigations, and discussions on how hallucination in prompts LLMs and even bolsters them in certain cases. | [
"Taveekitworachai, Pittawat",
"Abdullah, Febri",
"Thawonmas, Ruck"
] | Null-Shot Prompting: Rethinking Prompting Large Language Models With Hallucination | emnlp-main.740 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.741.bib | https://aclanthology.org/2024.emnlp-main.741/ | @inproceedings{naik-etal-2024-commvqa,
title = "{C}omm{VQA}: Situating Visual Question Answering in Communicative Contexts",
author = "Naik, Nandita Shankar and
Potts, Christopher and
Kreiss, Elisa",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.741",
pages = "13362--13377",
abstract = "Current visual question answering (VQA) models tend to be trained and evaluated on image-question pairs in isolation. However, the questions people ask are dependent on their informational needs and prior knowledge about the image content. To evaluate how situating images within naturalistic contexts shapes visual questions, we introduce CommVQA, a VQA dataset consisting of images, image descriptions, real-world communicative scenarios where the image might appear (e.g., a travel website), and follow-up questions and answers conditioned on the scenario and description. CommVQA, which contains 1000 images and 8,949 question-answer pairs, poses a challenge for current models. Error analyses and a human-subjects study suggest that generated answers still contain high rates of hallucinations, fail to fittingly address unanswerable questions, and don{'}t suitably reflect contextual information.",
}
| Current visual question answering (VQA) models tend to be trained and evaluated on image-question pairs in isolation. However, the questions people ask are dependent on their informational needs and prior knowledge about the image content. To evaluate how situating images within naturalistic contexts shapes visual questions, we introduce CommVQA, a VQA dataset consisting of images, image descriptions, real-world communicative scenarios where the image might appear (e.g., a travel website), and follow-up questions and answers conditioned on the scenario and description. CommVQA, which contains 1000 images and 8,949 question-answer pairs, poses a challenge for current models. Error analyses and a human-subjects study suggest that generated answers still contain high rates of hallucinations, fail to fittingly address unanswerable questions, and don{'}t suitably reflect contextual information. | [
"Naik, N",
"ita Shankar",
"Potts, Christopher",
"Kreiss, Elisa"
] | CommVQA: Situating Visual Question Answering in Communicative Contexts | emnlp-main.741 | Poster | 2402.15002 | [
"https://github.com/nnaik39/commvqa"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.742.bib | https://aclanthology.org/2024.emnlp-main.742/ | @inproceedings{zhao-etal-2024-ouroboros,
title = "Ouroboros: Generating Longer Drafts Phrase by Phrase for Faster Speculative Decoding",
author = "Zhao, Weilin and
Huang, Yuxiang and
Han, Xu and
Xu, Wang and
Xiao, Chaojun and
Zhang, Xinrong and
Fang, Yewei and
Zhang, Kaihuo and
Liu, Zhiyuan and
Sun, Maosong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.742",
pages = "13378--13393",
abstract = "Speculative decoding is a widely used method that accelerates the generation process of large language models (LLMs) with no compromise in model performance. It achieves this goal by using an existing smaller model for drafting and then employing the target LLM to verify the draft in a low-cost parallel manner. Under such a drafting-verification framework, drafting efficiency has become a bottleneck in the final speedup of speculative decoding. Therefore, generating longer drafts at less cost can lead to better decoding speedup. To achieve this, we introduce Ouroboros, which can generate draft phrases to parallelize the drafting process and meanwhile lengthen drafts in a training-free manner. The experimental results on various typical text generation tasks show that Ouroboros can achieve speedups of up to $2.4\times$ over speculative decoding and $3.9\times$ over vanilla decoding, without fine-tuning draft and target models. Code available at https://github.com/thunlp/Ouroboros.",
}
| Speculative decoding is a widely used method that accelerates the generation process of large language models (LLMs) with no compromise in model performance. It achieves this goal by using an existing smaller model for drafting and then employing the target LLM to verify the draft in a low-cost parallel manner. Under such a drafting-verification framework, drafting efficiency has become a bottleneck in the final speedup of speculative decoding. Therefore, generating longer drafts at less cost can lead to better decoding speedup. To achieve this, we introduce Ouroboros, which can generate draft phrases to parallelize the drafting process and meanwhile lengthen drafts in a training-free manner. The experimental results on various typical text generation tasks show that Ouroboros can achieve speedups of up to $2.4\times$ over speculative decoding and $3.9\times$ over vanilla decoding, without fine-tuning draft and target models. Code available at https://github.com/thunlp/Ouroboros. | [
"Zhao, Weilin",
"Huang, Yuxiang",
"Han, Xu",
"Xu, Wang",
"Xiao, Chaojun",
"Zhang, Xinrong",
"Fang, Yewei",
"Zhang, Kaihuo",
"Liu, Zhiyuan",
"Sun, Maosong"
] | Ouroboros: Generating Longer Drafts Phrase by Phrase for Faster Speculative Decoding | emnlp-main.742 | Poster | 2402.13720 | [
"https://github.com/thunlp/ouroboros"
] | https://huggingface.co/papers/2402.13720 | 1 | 5 | 1 | 6 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.743.bib | https://aclanthology.org/2024.emnlp-main.743/ | @inproceedings{huang-etal-2024-1,
title = "1+1{\textgreater}2: Can Large Language Models Serve as Cross-Lingual Knowledge Aggregators?",
author = "Huang, Yue and
Fan, Chenrui and
Li, Yuan and
Wu, Siyuan and
Zhou, Tianyi and
Zhang, Xiangliang and
Sun, Lichao",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.743",
pages = "13394--13412",
abstract = "Large Language Models (LLMs) have garnered significant attention due to their remarkable ability to process information across various languages. Despite their capabilities, they exhibit inconsistencies in handling identical queries in different languages, presenting challenges for further advancement. This paper introduces a method to enhance the multilingual performance of LLMs by aggregating knowledge from diverse languages. This approach incorporates a low-resource knowledge detector specific to a language, a strategic language selection process, and mechanisms for answer replacement and integration. Our extensive experiments demonstrate notable performance improvements, particularly in reducing the performance disparity across languages. An ablation study confirms that each component of our method significantly contributes to these enhancements. This research highlights the inherent potential of LLMs to harmonize multilingual capabilities and offers valuable insights for further exploration.",
}
| Large Language Models (LLMs) have garnered significant attention due to their remarkable ability to process information across various languages. Despite their capabilities, they exhibit inconsistencies in handling identical queries in different languages, presenting challenges for further advancement. This paper introduces a method to enhance the multilingual performance of LLMs by aggregating knowledge from diverse languages. This approach incorporates a low-resource knowledge detector specific to a language, a strategic language selection process, and mechanisms for answer replacement and integration. Our extensive experiments demonstrate notable performance improvements, particularly in reducing the performance disparity across languages. An ablation study confirms that each component of our method significantly contributes to these enhancements. This research highlights the inherent potential of LLMs to harmonize multilingual capabilities and offers valuable insights for further exploration. | [
"Huang, Yue",
"Fan, Chenrui",
"Li, Yuan",
"Wu, Siyuan",
"Zhou, Tianyi",
"Zhang, Xiangliang",
"Sun, Lichao"
] | 1+12: Can Large Language Models Serve as Cross-Lingual Knowledge Aggregators? | emnlp-main.743 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.744.bib | https://aclanthology.org/2024.emnlp-main.744/ | @inproceedings{xiao-etal-2024-leverage,
title = "How to Leverage Demonstration Data in Alignment for Large Language Model? A Self-Imitation Learning Perspective",
author = "Xiao, Teng and
Li, Mingxiao and
Yuan, Yige and
Zhu, Huaisheng and
Cui, Chao and
Honavar, Vasant G",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.744",
pages = "13413--13426",
abstract = "This paper introduces a novel generalized self-imitation learning GSIL framework, which effectively and efficiently aligns large language models with offline demonstration data. We develop GSIL by deriving a surrogate objective of imitation learning with density ratio estimates, facilitating the use of self-generated data and optimizing the imitation learning objective with simple classification losses. GSIL eliminates the need for complex adversarial training in standard imitation learning, achieving lightweight and efficient fine-tuning for large language models. In addition, GSIL encompasses a family of offline losses parameterized by a general class of convex functions for density ratio estimation and enables a unified view for alignment with demonstration data. Extensive experiments show that GSIL consistently and significantly outperforms baselines in many challenging benchmarks, such as coding (HuamnEval), mathematical reasoning (GSM8K) and instruction-following benchmark (MT-Bench). Code is public available at https://github.com/tengxiao1/GSIL.",
}
| This paper introduces a novel generalized self-imitation learning GSIL framework, which effectively and efficiently aligns large language models with offline demonstration data. We develop GSIL by deriving a surrogate objective of imitation learning with density ratio estimates, facilitating the use of self-generated data and optimizing the imitation learning objective with simple classification losses. GSIL eliminates the need for complex adversarial training in standard imitation learning, achieving lightweight and efficient fine-tuning for large language models. In addition, GSIL encompasses a family of offline losses parameterized by a general class of convex functions for density ratio estimation and enables a unified view for alignment with demonstration data. Extensive experiments show that GSIL consistently and significantly outperforms baselines in many challenging benchmarks, such as coding (HuamnEval), mathematical reasoning (GSM8K) and instruction-following benchmark (MT-Bench). Code is public available at https://github.com/tengxiao1/GSIL. | [
"Xiao, Teng",
"Li, Mingxiao",
"Yuan, Yige",
"Zhu, Huaisheng",
"Cui, Chao",
"Honavar, Vasant G"
] | How to Leverage Demonstration Data in Alignment for Large Language Model? A Self-Imitation Learning Perspective | emnlp-main.744 | Poster | 2410.10093 | [
"https://github.com/tengxiao1/gsil"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.745.bib | https://aclanthology.org/2024.emnlp-main.745/ | @inproceedings{lai-etal-2024-style,
title = "Style-Specific Neurons for Steering {LLM}s in Text Style Transfer",
author = "Lai, Wen and
Hangya, Viktor and
Fraser, Alexander",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.745",
pages = "13427--13443",
abstract = "Text style transfer (TST) aims to modify the style of a text without altering its original meaning. Large language models (LLMs) demonstrate superior performance across multiple tasks, including TST. However, in zero-shot setups, they tend to directly copy a significant portion of the input text to the output without effectively changing its style. To enhance the stylistic variety and fluency of the text, we present sNeuron-TST, a novel approach for steering LLMs using style-specific neurons in TST. Specifically, we identify neurons associated with the source and target styles and deactivate source-style-only neurons to give target-style words a higher probability, aiming to enhance the stylistic diversity of the generated text. However, we find that this deactivation negatively impacts the fluency of the generated text, which we address by proposing an improved contrastive decoding method that accounts for rapid token probability shifts across layers caused by deactivated source-style neurons. Empirical experiments demonstrate the effectiveness of the proposed method on six benchmarks, encompassing formality, toxicity, politics, politeness, authorship, and sentiment.",
}
| Text style transfer (TST) aims to modify the style of a text without altering its original meaning. Large language models (LLMs) demonstrate superior performance across multiple tasks, including TST. However, in zero-shot setups, they tend to directly copy a significant portion of the input text to the output without effectively changing its style. To enhance the stylistic variety and fluency of the text, we present sNeuron-TST, a novel approach for steering LLMs using style-specific neurons in TST. Specifically, we identify neurons associated with the source and target styles and deactivate source-style-only neurons to give target-style words a higher probability, aiming to enhance the stylistic diversity of the generated text. However, we find that this deactivation negatively impacts the fluency of the generated text, which we address by proposing an improved contrastive decoding method that accounts for rapid token probability shifts across layers caused by deactivated source-style neurons. Empirical experiments demonstrate the effectiveness of the proposed method on six benchmarks, encompassing formality, toxicity, politics, politeness, authorship, and sentiment. | [
"Lai, Wen",
"Hangya, Viktor",
"Fraser, Alex",
"er"
] | Style-Specific Neurons for Steering LLMs in Text Style Transfer | emnlp-main.745 | Poster | 2410.00593 | [
"https://github.com/wenlai-lavine/sneuron-tst"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.746.bib | https://aclanthology.org/2024.emnlp-main.746/ | @inproceedings{zhang-etal-2024-adaptive,
title = "Adaptive Query Rewriting: Aligning Rewriters through Marginal Probability of Conversational Answers",
author = "Zhang, Tianhua and
Li, Kun and
Luo, Hongyin and
Wu, Xixin and
Glass, James R. and
Meng, Helen M.",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.746",
pages = "13444--13461",
abstract = "Query rewriting is a crucial technique for passage retrieval in open-domain conversational question answering (CQA). It decontexualizes conversational queries into self-contained questions suitable for off-the-shelf retrievers. Existing methods attempt to incorporate retriever{'}s preference during the training of rewriting models. However, these approaches typically rely on extensive annotations such as in-domain rewrites and/or relevant passage labels, limiting the models{'} generalization and adaptation capabilities. In this paper, we introduce AdaQR (Adaptive Query Rewriting), a framework for training query rewriting models with limited rewrite annotations from seed datasets and completely no passage label. Our approach begins by fine-tuning compact large language models using only 10{\%} of rewrite annotations from the seed dataset training split. The models are then utilized to self-sample rewrite candidates for each query instance, further eliminating the expense for human labeling or larger language model prompting often adopted in curating preference data. A novel approach is then proposed to assess retriever{'}s preference for these candidates with the probability of answers conditioned on the conversational query by marginalizing the Top-$K$ passages. This serves as the reward for optimizing the rewriter further using Direct Preference Optimization (DPO), a process free of rewrite and retrieval annotations. Experimental results on four open-domain CQA datasets demonstrate that AdaQR not only enhances the in-domain capabilities of the rewriter with limited annotation requirement, but also adapts effectively to out-of-domain datasets.",
}
| Query rewriting is a crucial technique for passage retrieval in open-domain conversational question answering (CQA). It decontexualizes conversational queries into self-contained questions suitable for off-the-shelf retrievers. Existing methods attempt to incorporate retriever{'}s preference during the training of rewriting models. However, these approaches typically rely on extensive annotations such as in-domain rewrites and/or relevant passage labels, limiting the models{'} generalization and adaptation capabilities. In this paper, we introduce AdaQR (Adaptive Query Rewriting), a framework for training query rewriting models with limited rewrite annotations from seed datasets and completely no passage label. Our approach begins by fine-tuning compact large language models using only 10{\%} of rewrite annotations from the seed dataset training split. The models are then utilized to self-sample rewrite candidates for each query instance, further eliminating the expense for human labeling or larger language model prompting often adopted in curating preference data. A novel approach is then proposed to assess retriever{'}s preference for these candidates with the probability of answers conditioned on the conversational query by marginalizing the Top-$K$ passages. This serves as the reward for optimizing the rewriter further using Direct Preference Optimization (DPO), a process free of rewrite and retrieval annotations. Experimental results on four open-domain CQA datasets demonstrate that AdaQR not only enhances the in-domain capabilities of the rewriter with limited annotation requirement, but also adapts effectively to out-of-domain datasets. | [
"Zhang, Tianhua",
"Li, Kun",
"Luo, Hongyin",
"Wu, Xixin",
"Glass, James R.",
"Meng, Helen M."
] | Adaptive Query Rewriting: Aligning Rewriters through Marginal Probability of Conversational Answers | emnlp-main.746 | Oral | 2406.10991 | [
""
] | https://huggingface.co/papers/2406.10991 | 0 | 0 | 1 | 6 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.747.bib | https://aclanthology.org/2024.emnlp-main.747/ | @inproceedings{zhou-etal-2024-grasping,
title = "Grasping the Essentials: Tailoring Large Language Models for Zero-Shot Relation Extraction",
author = "Zhou, Sizhe and
Meng, Yu and
Jin, Bowen and
Han, Jiawei",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.747",
pages = "13462--13486",
abstract = "Relation extraction (RE) aims to identify semantic relationships between entities within text. Despite considerable advancements, existing models predominantly require extensive annotated training data, which is both costly and labor-intensive to collect. Moreover, these models often struggle to adapt to new or unseen relations. Few-shot learning, aiming to lessen annotation demands, typically provides incomplete and biased supervision for target relations, leading to degraded and unstable performance. To accurately and explicitly describe relation semantics while minimizing annotation demands, we explore the definition only zero-shot RE setting where only relation definitions expressed in natural language are used to train a RE model. We introduce REPaL, comprising three stages: (1) We leverage large language models (LLMs) to generate initial seed instances from relation definitions and an unlabeled corpus. (2) We fine-tune a bidirectional Small Language Model (SLM) with initial seeds to learn relations for the target domain. (3) We expand pattern coverage and mitigate bias from initial seeds by integrating feedback from the SLM{'}s predictions on the unlabeled corpus and the synthesis history. To accomplish this, we leverage the multi-turn conversation ability of LLMs to generate new instances in follow-up dialogues, informed by both the feedback and synthesis history. Studies reveal that definition-oriented seed synthesis enhances pattern coverage whereas indiscriminately increasing seed quantity leads to performance saturation. Experiments on two datasets show REPaL significantly improved cost-effective zero-shot performance by large margins.",
}
| Relation extraction (RE) aims to identify semantic relationships between entities within text. Despite considerable advancements, existing models predominantly require extensive annotated training data, which is both costly and labor-intensive to collect. Moreover, these models often struggle to adapt to new or unseen relations. Few-shot learning, aiming to lessen annotation demands, typically provides incomplete and biased supervision for target relations, leading to degraded and unstable performance. To accurately and explicitly describe relation semantics while minimizing annotation demands, we explore the definition only zero-shot RE setting where only relation definitions expressed in natural language are used to train a RE model. We introduce REPaL, comprising three stages: (1) We leverage large language models (LLMs) to generate initial seed instances from relation definitions and an unlabeled corpus. (2) We fine-tune a bidirectional Small Language Model (SLM) with initial seeds to learn relations for the target domain. (3) We expand pattern coverage and mitigate bias from initial seeds by integrating feedback from the SLM{'}s predictions on the unlabeled corpus and the synthesis history. To accomplish this, we leverage the multi-turn conversation ability of LLMs to generate new instances in follow-up dialogues, informed by both the feedback and synthesis history. Studies reveal that definition-oriented seed synthesis enhances pattern coverage whereas indiscriminately increasing seed quantity leads to performance saturation. Experiments on two datasets show REPaL significantly improved cost-effective zero-shot performance by large margins. | [
"Zhou, Sizhe",
"Meng, Yu",
"Jin, Bowen",
"Han, Jiawei"
] | Grasping the Essentials: Tailoring Large Language Models for Zero-Shot Relation Extraction | emnlp-main.747 | Poster | 2402.11142 | [
"https://github.com/kevinsrr/repal"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.748.bib | https://aclanthology.org/2024.emnlp-main.748/ | @inproceedings{huang-etal-2024-da,
title = "{DA}-Code: Agent Data Science Code Generation Benchmark for Large Language Models",
author = "Huang, Yiming and
Luo, Jianwen and
Yu, Yan and
Zhang, Yitong and
Lei, Fangyu and
Wei, Yifan and
He, Shizhu and
Huang, Lifu and
Liu, Xiao and
Zhao, Jun and
Liu, Kang",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.748",
pages = "13487--13521",
abstract = "We introduce DA-Code, a code generation benchmark specifically designed to assess LLMs on agent-based data science tasks. This benchmark features three core elements: First, the tasks within DA-Code are inherently challenging, setting them apart from traditional code generation tasks and demanding advanced coding skills in grounding and planning. Second, examples in DA-Code are all based on real and diverse data, covering a wide range of complex data wrangling and analytics tasks. Third, to solve the tasks, the models must utilize complex data science programming languages, including Python and SQL, to perform intricate data processing and derive the answers. We set up the benchmark in a controllable and executable environment that aligns with real-world data analysis scenarios and is scalable. The annotators meticulously designed the evaluation suite to ensure the accuracy and robustness of the evaluation. We developed the DA-Agent baseline. Experiments show that although the baseline performs better than other existing frameworks, using the current best LLMs achieves only 30.5{\%} accuracy, leaving ample room for improvement. We release our benchmark at [link](https://github.com/yiyihum/dabench)",
}
| We introduce DA-Code, a code generation benchmark specifically designed to assess LLMs on agent-based data science tasks. This benchmark features three core elements: First, the tasks within DA-Code are inherently challenging, setting them apart from traditional code generation tasks and demanding advanced coding skills in grounding and planning. Second, examples in DA-Code are all based on real and diverse data, covering a wide range of complex data wrangling and analytics tasks. Third, to solve the tasks, the models must utilize complex data science programming languages, including Python and SQL, to perform intricate data processing and derive the answers. We set up the benchmark in a controllable and executable environment that aligns with real-world data analysis scenarios and is scalable. The annotators meticulously designed the evaluation suite to ensure the accuracy and robustness of the evaluation. We developed the DA-Agent baseline. Experiments show that although the baseline performs better than other existing frameworks, using the current best LLMs achieves only 30.5{\%} accuracy, leaving ample room for improvement. We release our benchmark at [link](https://github.com/yiyihum/dabench) | [
"Huang, Yiming",
"Luo, Jianwen",
"Yu, Yan",
"Zhang, Yitong",
"Lei, Fangyu",
"Wei, Yifan",
"He, Shizhu",
"Huang, Lifu",
"Liu, Xiao",
"Zhao, Jun",
"Liu, Kang"
] | DA-Code: Agent Data Science Code Generation Benchmark for Large Language Models | emnlp-main.748 | Poster | 2410.07331 | [
""
] | https://huggingface.co/papers/2410.07331 | 4 | 4 | 3 | 11 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.749.bib | https://aclanthology.org/2024.emnlp-main.749/ | @inproceedings{jiang-etal-2024-leveraging-context,
title = "Leveraging Context-Aware Prompting for Commit Message Generation",
author = "Jiang, Zhihua and
Chen, Jianwei and
Rao, Dongning and
Ye, Guanghui",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.749",
pages = "13522--13540",
abstract = "Writing comprehensive commit messages is tedious yet important, because these messages describe changes of code, such as fixing bugs or adding new features. However, most existing methods focus on either only the changed lines or nearest context lines, without considering the effectiveness of selecting useful contexts. On the other hand, it is possible that introducing excessive contexts can lead to noise. To this end, we propose a code model COMMIT (Context-aware prOMpting based comMIt-message generaTion) in conjunction with a code dataset CODEC (COntext and metaData Enhanced Code dataset). Leveraging program slicing, CODEC consolidates code changes along with related contexts via property graph analysis. Further, utilizing CodeT5+ as the backbone model, we train COMMIT via context-aware prompt on CODEC. Experiments show that COMMIT can surpass all compared models including pre-trained language models for code (code-PLMs) such as CommitBART and large language models for code (code-LLMs) such as Code-LlaMa. Besides, we investigate several research questions (RQs), further verifying the effectiveness of our approach. We release the data and code at: https://github.com/Jnunlplab/COMMIT.git.",
}
| Writing comprehensive commit messages is tedious yet important, because these messages describe changes of code, such as fixing bugs or adding new features. However, most existing methods focus on either only the changed lines or nearest context lines, without considering the effectiveness of selecting useful contexts. On the other hand, it is possible that introducing excessive contexts can lead to noise. To this end, we propose a code model COMMIT (Context-aware prOMpting based comMIt-message generaTion) in conjunction with a code dataset CODEC (COntext and metaData Enhanced Code dataset). Leveraging program slicing, CODEC consolidates code changes along with related contexts via property graph analysis. Further, utilizing CodeT5+ as the backbone model, we train COMMIT via context-aware prompt on CODEC. Experiments show that COMMIT can surpass all compared models including pre-trained language models for code (code-PLMs) such as CommitBART and large language models for code (code-LLMs) such as Code-LlaMa. Besides, we investigate several research questions (RQs), further verifying the effectiveness of our approach. We release the data and code at: https://github.com/Jnunlplab/COMMIT.git. | [
"Jiang, Zhihua",
"Chen, Jianwei",
"Rao, Dongning",
"Ye, Guanghui"
] | Leveraging Context-Aware Prompting for Commit Message Generation | emnlp-main.749 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.750.bib | https://aclanthology.org/2024.emnlp-main.750/ | @inproceedings{fleisig-etal-2024-linguistic,
title = "Linguistic Bias in {C}hat{GPT}: Language Models Reinforce Dialect Discrimination",
author = "Fleisig, Eve and
Smith, Genevieve and
Bossi, Madeline and
Rustagi, Ishita and
Yin, Xavier and
Klein, Dan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.750",
pages = "13541--13564",
abstract = "We present a large-scale study of linguistic bias exhibited by ChatGPT covering ten dialects of English (Standard American English, Standard British English, and eight widely spoken non-{''}standard{''} varieties from around the world). We prompted GPT-3.5 Turbo and GPT-4 with text by native speakers of each variety and analyzed the responses via detailed linguistic feature annotation and native speaker evaluation. We find that the models default to {``}standard{''} varieties of English; based on evaluation by native speakers, we also find that model responses to non-{''}standard{''} varieties consistently exhibit a range of issues: stereotyping (19{\%} worse than for {``}standard{''} varieties), demeaning content (25{\%} worse), lack of comprehension (9{\%} worse), and condescending responses (15{\%} worse). Moreover, if these models are asked to imitate the writing style of prompts in non-{''}standard{''} varieties, they produce text that exhibits lower comprehension of the input and is especially prone to stereotyping. GPT-4 improves on GPT-3.5 in terms of comprehension, warmth, and friendliness, but also exhibits a marked increase in stereotyping (+18{\%}). The results indicate that GPT-3.5 Turbo and GPT-4 can perpetuate linguistic discrimination toward speakers of non-{''}standard{''} varieties.",
}
| We present a large-scale study of linguistic bias exhibited by ChatGPT covering ten dialects of English (Standard American English, Standard British English, and eight widely spoken non-{''}standard{''} varieties from around the world). We prompted GPT-3.5 Turbo and GPT-4 with text by native speakers of each variety and analyzed the responses via detailed linguistic feature annotation and native speaker evaluation. We find that the models default to {``}standard{''} varieties of English; based on evaluation by native speakers, we also find that model responses to non-{''}standard{''} varieties consistently exhibit a range of issues: stereotyping (19{\%} worse than for {``}standard{''} varieties), demeaning content (25{\%} worse), lack of comprehension (9{\%} worse), and condescending responses (15{\%} worse). Moreover, if these models are asked to imitate the writing style of prompts in non-{''}standard{''} varieties, they produce text that exhibits lower comprehension of the input and is especially prone to stereotyping. GPT-4 improves on GPT-3.5 in terms of comprehension, warmth, and friendliness, but also exhibits a marked increase in stereotyping (+18{\%}). The results indicate that GPT-3.5 Turbo and GPT-4 can perpetuate linguistic discrimination toward speakers of non-{''}standard{''} varieties. | [
"Fleisig, Eve",
"Smith, Genevieve",
"Bossi, Madeline",
"Rustagi, Ishita",
"Yin, Xavier",
"Klein, Dan"
] | Linguistic Bias in ChatGPT: Language Models Reinforce Dialect Discrimination | emnlp-main.750 | Poster | 2406.08818 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.751.bib | https://aclanthology.org/2024.emnlp-main.751/ | @inproceedings{chen-etal-2024-lifelong,
title = "Lifelong Knowledge Editing for {LLM}s with Retrieval-Augmented Continuous Prompt Learning",
author = "Chen, Qizhou and
Zhang, Taolin and
He, Xiaofeng and
Li, Dongyang and
Wang, Chengyu and
Huang, Longtao and
Xue{'}, Hui",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.751",
pages = "13565--13580",
abstract = "Model editing aims to correct outdated or erroneous knowledge in large language models (LLMs) without the need for costly retraining. Lifelong model editing is the most challenging task that caters to the continuous editing requirements of LLMs. Prior works primarily focus on single or batch editing; nevertheless, these methods fall short in lifelong editing scenarios due to catastrophic knowledge forgetting and the degradation of model performance. Although retrieval-based methods alleviate these issues, they are impeded by slow and cumbersome processes of integrating the retrieved knowledge into the model. In this work, we introduce RECIPE, a RetriEval-augmented ContInuous Prompt lEarning method, to boost editing efficacy and inference efficiency in lifelong learning. RECIPE first converts knowledge statements into short and informative continuous prompts, prefixed to the LLM{'}s input query embedding, to efficiently refine the response grounded on the knowledge. It further integrates the Knowledge Sentinel (KS) that acts as an intermediary to calculate a dynamic threshold, determining whether the retrieval repository contains relevant knowledge. Our retriever and prompt encoder are jointly trained to achieve editing properties, i.e., reliability, generality, and locality. In our experiments, RECIPE is assessed extensively across multiple LLMs and editing datasets, where it achieves superior editing performance. RECIPE also demonstrates its capability to maintain the overall performance of LLMs alongside showcasing fast editing and inference speed.",
}
| Model editing aims to correct outdated or erroneous knowledge in large language models (LLMs) without the need for costly retraining. Lifelong model editing is the most challenging task that caters to the continuous editing requirements of LLMs. Prior works primarily focus on single or batch editing; nevertheless, these methods fall short in lifelong editing scenarios due to catastrophic knowledge forgetting and the degradation of model performance. Although retrieval-based methods alleviate these issues, they are impeded by slow and cumbersome processes of integrating the retrieved knowledge into the model. In this work, we introduce RECIPE, a RetriEval-augmented ContInuous Prompt lEarning method, to boost editing efficacy and inference efficiency in lifelong learning. RECIPE first converts knowledge statements into short and informative continuous prompts, prefixed to the LLM{'}s input query embedding, to efficiently refine the response grounded on the knowledge. It further integrates the Knowledge Sentinel (KS) that acts as an intermediary to calculate a dynamic threshold, determining whether the retrieval repository contains relevant knowledge. Our retriever and prompt encoder are jointly trained to achieve editing properties, i.e., reliability, generality, and locality. In our experiments, RECIPE is assessed extensively across multiple LLMs and editing datasets, where it achieves superior editing performance. RECIPE also demonstrates its capability to maintain the overall performance of LLMs alongside showcasing fast editing and inference speed. | [
"Chen, Qizhou",
"Zhang, Taolin",
"He, Xiaofeng",
"Li, Dongyang",
"Wang, Chengyu",
"Huang, Longtao",
"Xue{'}, Hui"
] | Lifelong Knowledge Editing for LLMs with Retrieval-Augmented Continuous Prompt Learning | emnlp-main.751 | Poster | 2405.03279 | [
"https://github.com/qizhou000/RECIPE"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.752.bib | https://aclanthology.org/2024.emnlp-main.752/ | @inproceedings{wang-etal-2024-learning-rate,
title = "A Learning Rate Path Switching Training Paradigm for Version Updates of Large Language Models",
author = "Wang, Zhihao and
Liu, Shiyu and
Huang, Jianheng and
Zheng, Wang and
Liao, YiXuan and
Chen, Xiaoxin and
Yao, Junfeng and
Su, Jinsong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.752",
pages = "13581--13594",
abstract = "Due to the continuous emergence of new data, version updates have become an indispensable requirement for Large Language Models (LLMs). The training paradigms for version updates of LLMs include pre-training from scratch (PTFS) and continual pre-training (CPT). Preliminary experiments demonstrate that PTFS achieves better pre-training performance, while CPT has lower training cost. Moreover, their performance and training cost gaps widen progressively with version updates. To investigate the underlying reasons for this phenomenon, we analyze the effect of learning rate adjustments during the two stages of CPT: preparing an initialization checkpoint and continual pre-training based on this checkpoint. We find that a large learning rate in the first stage and a complete learning rate decay process in the second stage are crucial for version updates of LLMs. Hence, we propose a learning rate path switching training paradigm. Our paradigm comprises one main path, where we pre-train a LLM with the maximal learning rate, and multiple branching paths, each of which corresponds to an update of the LLM with newly-added training data. Extensive experiments demonstrate the effectiveness and generalization of our paradigm. Particularly, when training four versions of LLMs, our paradigm reduces the total training cost to 58{\%} compared to PTFS, while maintaining comparable pre-training performance.",
}
| Due to the continuous emergence of new data, version updates have become an indispensable requirement for Large Language Models (LLMs). The training paradigms for version updates of LLMs include pre-training from scratch (PTFS) and continual pre-training (CPT). Preliminary experiments demonstrate that PTFS achieves better pre-training performance, while CPT has lower training cost. Moreover, their performance and training cost gaps widen progressively with version updates. To investigate the underlying reasons for this phenomenon, we analyze the effect of learning rate adjustments during the two stages of CPT: preparing an initialization checkpoint and continual pre-training based on this checkpoint. We find that a large learning rate in the first stage and a complete learning rate decay process in the second stage are crucial for version updates of LLMs. Hence, we propose a learning rate path switching training paradigm. Our paradigm comprises one main path, where we pre-train a LLM with the maximal learning rate, and multiple branching paths, each of which corresponds to an update of the LLM with newly-added training data. Extensive experiments demonstrate the effectiveness and generalization of our paradigm. Particularly, when training four versions of LLMs, our paradigm reduces the total training cost to 58{\%} compared to PTFS, while maintaining comparable pre-training performance. | [
"Wang, Zhihao",
"Liu, Shiyu",
"Huang, Jianheng",
"Zheng, Wang",
"Liao, YiXuan",
"Chen, Xiaoxin",
"Yao, Junfeng",
"Su, Jinsong"
] | A Learning Rate Path Switching Training Paradigm for Version Updates of Large Language Models | emnlp-main.752 | Poster | 2410.04103 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.753.bib | https://aclanthology.org/2024.emnlp-main.753/ | @inproceedings{sohn-etal-2024-zero,
title = "Zero-Shot Cross-Lingual {NER} Using Phonemic Representations for Low-Resource Languages",
author = "Sohn, Jimin and
Jung, Haeji and
Cheng, Alex and
Kang, Jooeon and
Du, Yilin and
Mortensen, David R",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.753",
pages = "13595--13602",
abstract = "Existing zero-shot cross-lingual NER approaches require substantial prior knowledge of the target language, which is impractical for low-resource languages.In this paper, we propose a novel approach to NER using phonemic representation based on the International Phonetic Alphabet (IPA) to bridge the gap between representations of different languages.Our experiments show that our method significantly outperforms baseline models in extremely low-resource languages, with the highest average F1 score (46.38{\%}) and lowest standard deviation (12.67), particularly demonstrating its robustness with non-Latin scripts. Ourcodes are available at https://github.com/Gabriel819/zeroshot{\_}ner.git",
}
| Existing zero-shot cross-lingual NER approaches require substantial prior knowledge of the target language, which is impractical for low-resource languages.In this paper, we propose a novel approach to NER using phonemic representation based on the International Phonetic Alphabet (IPA) to bridge the gap between representations of different languages.Our experiments show that our method significantly outperforms baseline models in extremely low-resource languages, with the highest average F1 score (46.38{\%}) and lowest standard deviation (12.67), particularly demonstrating its robustness with non-Latin scripts. Ourcodes are available at https://github.com/Gabriel819/zeroshot{\_}ner.git | [
"Sohn, Jimin",
"Jung, Haeji",
"Cheng, Alex",
"Kang, Jooeon",
"Du, Yilin",
"Mortensen, David R"
] | Zero-Shot Cross-Lingual NER Using Phonemic Representations for Low-Resource Languages | emnlp-main.753 | Poster | 2406.16030 | [
"https://github.com/gabriel819/zeroshot_ner"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.754.bib | https://aclanthology.org/2024.emnlp-main.754/ | @inproceedings{lv-etal-2024-analysis,
title = "An Analysis and Mitigation of the Reversal Curse",
author = "Lv, Ang and
Zhang, Kaiyi and
Xie, Shufang and
Tu, Quan and
Chen, Yuhan and
Wen, Ji-Rong and
Yan, Rui",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.754",
pages = "13603--13615",
}
| No abstract found | [
"Lv, Ang",
"Zhang, Kaiyi",
"Xie, Shufang",
"Tu, Quan",
"Chen, Yuhan",
"Wen, Ji-Rong",
"Yan, Rui"
] | An Analysis and Mitigation of the Reversal Curse | emnlp-main.754 | Poster | 2311.07468 | [
"https://github.com/trestad/mitigating-reversal-curse"
] | https://huggingface.co/papers/2311.07468 | 0 | 1 | 0 | 7 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.755.bib | https://aclanthology.org/2024.emnlp-main.755/ | @inproceedings{kim-etal-2024-exploring-practicality,
title = "Exploring the Practicality of Generative Retrieval on Dynamic Corpora",
author = "Kim, Chaeeun and
Yoon, Soyoung and
Lee, Hyunji and
Jang, Joel and
Yang, Sohee and
Seo, Minjoon",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.755",
pages = "13616--13633",
abstract = "Benchmarking the performance of information retrieval (IR) is mostly conducted with a fixed set of documents (static corpora). However, in realistic scenarios, this is rarely the case and the documents to be retrieved are constantly updated and added. In this paper, we focus on Generative Retrievals (GR), which apply autoregressive language models to IR problems, and explore their adaptability and robustness in dynamic scenarios. We also conduct an extensive evaluation of computational and memory efficiency, crucial factors for real-world deployment of IR systems handling vast and ever-changing document collections. Our results on the StreamingQA benchmark demonstrate that GR is more adaptable to evolving knowledge (4{--}11{\%}), robust in learning knowledge with temporal information, and efficient in terms of inference FLOPs (x2), indexing time (x6), and storage footprint (x4) compared to Dual Encoders (DE), which are commonly used in retrieval systems. Our paper highlights the potential of GR for future use in practical IR systems within dynamic environments.",
}
| Benchmarking the performance of information retrieval (IR) is mostly conducted with a fixed set of documents (static corpora). However, in realistic scenarios, this is rarely the case and the documents to be retrieved are constantly updated and added. In this paper, we focus on Generative Retrievals (GR), which apply autoregressive language models to IR problems, and explore their adaptability and robustness in dynamic scenarios. We also conduct an extensive evaluation of computational and memory efficiency, crucial factors for real-world deployment of IR systems handling vast and ever-changing document collections. Our results on the StreamingQA benchmark demonstrate that GR is more adaptable to evolving knowledge (4{--}11{\%}), robust in learning knowledge with temporal information, and efficient in terms of inference FLOPs (x2), indexing time (x6), and storage footprint (x4) compared to Dual Encoders (DE), which are commonly used in retrieval systems. Our paper highlights the potential of GR for future use in practical IR systems within dynamic environments. | [
"Kim, Chaeeun",
"Yoon, Soyoung",
"Lee, Hyunji",
"Jang, Joel",
"Yang, Sohee",
"Seo, Minjoon"
] | Exploring the Practicality of Generative Retrieval on Dynamic Corpora | emnlp-main.755 | Poster | 2305.18952 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.756.bib | https://aclanthology.org/2024.emnlp-main.756/ | @inproceedings{liu-etal-2024-onenet,
title = "{O}ne{N}et: A Fine-Tuning Free Framework for Few-Shot Entity Linking via Large Language Model Prompting",
author = "Liu, Xukai and
Liu, Ye and
Zhang, Kai and
Wang, Kehang and
Liu, Qi and
Chen, Enhong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.756",
pages = "13634--13651",
abstract = "Entity Linking (EL) is the process of associating ambiguous textual mentions to specific entities in a knowledge base.Traditional EL methods heavily rely on large datasets to enhance their performance, a dependency that becomes problematic in the context of few-shot entity linking, where only a limited number of examples are available for training. To address this challenge, we present OneNet, an innovative framework that utilizes the few-shot learning capabilities of Large Language Models (LLMs) without the need for fine-tuning. To the best of our knowledge, this marks a pioneering approach to applying LLMs to few-shot entity linking tasks. OneNet is structured around three key components prompted by LLMs: (1) an entity reduction processor that simplifies inputs by summarizing and filtering out irrelevant entities, (2) a dual-perspective entity linker that combines contextual cues and prior knowledge for precise entity linking, and (3) an entity consensus judger that employs a unique consistency algorithm to alleviate the hallucination in the entity linking reasoning.Comprehensive evaluations across seven benchmark datasets reveal that OneNet outperforms current state-of-the-art entity linking methods.",
}
| Entity Linking (EL) is the process of associating ambiguous textual mentions to specific entities in a knowledge base.Traditional EL methods heavily rely on large datasets to enhance their performance, a dependency that becomes problematic in the context of few-shot entity linking, where only a limited number of examples are available for training. To address this challenge, we present OneNet, an innovative framework that utilizes the few-shot learning capabilities of Large Language Models (LLMs) without the need for fine-tuning. To the best of our knowledge, this marks a pioneering approach to applying LLMs to few-shot entity linking tasks. OneNet is structured around three key components prompted by LLMs: (1) an entity reduction processor that simplifies inputs by summarizing and filtering out irrelevant entities, (2) a dual-perspective entity linker that combines contextual cues and prior knowledge for precise entity linking, and (3) an entity consensus judger that employs a unique consistency algorithm to alleviate the hallucination in the entity linking reasoning.Comprehensive evaluations across seven benchmark datasets reveal that OneNet outperforms current state-of-the-art entity linking methods. | [
"Liu, Xukai",
"Liu, Ye",
"Zhang, Kai",
"Wang, Kehang",
"Liu, Qi",
"Chen, Enhong"
] | OneNet: A Fine-Tuning Free Framework for Few-Shot Entity Linking via Large Language Model Prompting | emnlp-main.756 | Oral | 2410.07549 | [
"https://github.com/laquabe/OneNet"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.757.bib | https://aclanthology.org/2024.emnlp-main.757/ | @inproceedings{deng-etal-2024-dont,
title = "Don{'}t Just Say {``}{I} don{'}t know{''}! Self-aligning Large Language Models for Responding to Unknown Questions with Explanations",
author = "Deng, Yang and
Zhao, Yong and
Li, Moxin and
Ng, See-Kiong and
Chua, Tat-Seng",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.757",
pages = "13652--13673",
abstract = "Despite the remarkable abilities of Large Language Models (LLMs) to answer questions, they often display a considerable level of overconfidence even when the question does not have a definitive answer. To avoid providing hallucinated answers to these unknown questions, existing studies typically investigate approaches to refusing to answer these questions. In this work, we propose a novel and scalable self-alignment method to utilize the LLM itself to enhance its response-ability to different types of unknown questions, being capable of not just refusing to answer but further proactively providing explanations to the unanswerability of unknown questions. Specifically, the Self-Align method first employ a two-stage class-aware self-augmentation approach to generate a large amount of unknown question-response data. Then we conduct disparity-driven self-curation to select qualified data for fine-tuning the LLM itself for aligning the responses to unknown questions as desired. Experimental results on two datasets across four types of unknown questions validate the superiority of the Self-Aligned method over existing baselines in terms of three types of task formulation.",
}
| Despite the remarkable abilities of Large Language Models (LLMs) to answer questions, they often display a considerable level of overconfidence even when the question does not have a definitive answer. To avoid providing hallucinated answers to these unknown questions, existing studies typically investigate approaches to refusing to answer these questions. In this work, we propose a novel and scalable self-alignment method to utilize the LLM itself to enhance its response-ability to different types of unknown questions, being capable of not just refusing to answer but further proactively providing explanations to the unanswerability of unknown questions. Specifically, the Self-Align method first employ a two-stage class-aware self-augmentation approach to generate a large amount of unknown question-response data. Then we conduct disparity-driven self-curation to select qualified data for fine-tuning the LLM itself for aligning the responses to unknown questions as desired. Experimental results on two datasets across four types of unknown questions validate the superiority of the Self-Aligned method over existing baselines in terms of three types of task formulation. | [
"Deng, Yang",
"Zhao, Yong",
"Li, Moxin",
"Ng, See-Kiong",
"Chua, Tat-Seng"
] | Don't Just Say “I don't know”! Self-aligning Large Language Models for Responding to Unknown Questions with Explanations | emnlp-main.757 | Poster | [
"https://github.com/zhaoy777/kuqp-dataset"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.758.bib | https://aclanthology.org/2024.emnlp-main.758/ | @inproceedings{huang-etal-2024-fewer,
title = "Fewer is More: Boosting Math Reasoning with Reinforced Context Pruning",
author = "Huang, Xijie and
Zhang, Li Lyna and
Cheng, Kwang-Ting and
Yang, Fan and
Yang, Mao",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.758",
pages = "13674--13695",
abstract = "Large Language Models (LLMs) have shown impressive capabilities, yet they still struggle with math reasoning. In this work, we propose CoT-Influx, a novel approach that pushes the boundary of few-shot Chain-of-Thoughts (CoT) learning to improve LLM mathematical reasoning. Motivated by the observation that adding more concise CoT examples in the prompt can improve LLM reasoning performance, CoT-Influx employs a coarse-to-fine pruner to maximize the input of effective and concise CoT examples. The pruner first selects as many crucial CoT examples as possible and then prunes unimportant tokens to fit the context window. As a result, by enabling more CoT examples with double the context window size in tokens, CoT-Influx significantly outperforms various prompting baselines across various LLMs (LLaMA2-7B, 13B, 70B) and 5 math datasets, achieving up to 4.55{\%} absolute improvements. Remarkably, without any fine-tuning, LLaMA2-70B with CoT-Influx surpasses GPT-3.5 and a wide range of larger LLMs (PaLM, Minerva 540B, etc.) on the GSM8K. CoT-Influx is a plug-and-play module for LLMs, adaptable in various scenarios. It{'}s compatible with advanced reasoning prompting techniques, such as self-consistency, and supports different long-context LLMs, including Mistral-7B-v0.3-32K and Yi-6B-200K.",
}
| Large Language Models (LLMs) have shown impressive capabilities, yet they still struggle with math reasoning. In this work, we propose CoT-Influx, a novel approach that pushes the boundary of few-shot Chain-of-Thoughts (CoT) learning to improve LLM mathematical reasoning. Motivated by the observation that adding more concise CoT examples in the prompt can improve LLM reasoning performance, CoT-Influx employs a coarse-to-fine pruner to maximize the input of effective and concise CoT examples. The pruner first selects as many crucial CoT examples as possible and then prunes unimportant tokens to fit the context window. As a result, by enabling more CoT examples with double the context window size in tokens, CoT-Influx significantly outperforms various prompting baselines across various LLMs (LLaMA2-7B, 13B, 70B) and 5 math datasets, achieving up to 4.55{\%} absolute improvements. Remarkably, without any fine-tuning, LLaMA2-70B with CoT-Influx surpasses GPT-3.5 and a wide range of larger LLMs (PaLM, Minerva 540B, etc.) on the GSM8K. CoT-Influx is a plug-and-play module for LLMs, adaptable in various scenarios. It{'}s compatible with advanced reasoning prompting techniques, such as self-consistency, and supports different long-context LLMs, including Mistral-7B-v0.3-32K and Yi-6B-200K. | [
"Huang, Xijie",
"Zhang, Li Lyna",
"Cheng, Kwang-Ting",
"Yang, Fan",
"Yang, Mao"
] | Fewer is More: Boosting Math Reasoning with Reinforced Context Pruning | emnlp-main.758 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.759.bib | https://aclanthology.org/2024.emnlp-main.759/ | @inproceedings{liu-etal-2024-large,
title = "Large Language Models Are Poor Clinical Decision-Makers: A Comprehensive Benchmark",
author = "Liu, Fenglin and
Li, Zheng and
Zhou, Hongjian and
Yin, Qingyu and
Yang, Jingfeng and
Tang, Xianfeng and
Luo, Chen and
Zeng, Ming and
Jiang, Haoming and
Gao, Yifan and
Nigam, Priyanka and
Nag, Sreyashi and
Yin, Bing and
Hua, Yining and
Zhou, Xuan and
Rohanian, Omid and
Thakur, Anshul and
Clifton, Lei and
Clifton, David A.",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.759",
pages = "13696--13710",
abstract = "The adoption of large language models (LLMs) to assist clinicians has attracted remarkable attention. Existing works mainly adopt the close-ended question-answering (QA) task with answer options for evaluation. However, many clinical decisions involve answering open-ended questions without pre-set options. To better understand LLMs in the clinic, we construct a benchmark ClinicBench. We first collect eleven existing datasets covering diverse clinical language generation, understanding, and reasoning tasks. Furthermore, we construct six novel datasets and clinical tasks that are complex but common in real-world practice, e.g., open-ended decision-making, long document processing, and emerging drug analysis. We conduct an extensive evaluation of twenty-two LLMs under both zero-shot and few-shot settings. Finally, we invite medical experts to evaluate the clinical usefulness of LLMs",
}
| The adoption of large language models (LLMs) to assist clinicians has attracted remarkable attention. Existing works mainly adopt the close-ended question-answering (QA) task with answer options for evaluation. However, many clinical decisions involve answering open-ended questions without pre-set options. To better understand LLMs in the clinic, we construct a benchmark ClinicBench. We first collect eleven existing datasets covering diverse clinical language generation, understanding, and reasoning tasks. Furthermore, we construct six novel datasets and clinical tasks that are complex but common in real-world practice, e.g., open-ended decision-making, long document processing, and emerging drug analysis. We conduct an extensive evaluation of twenty-two LLMs under both zero-shot and few-shot settings. Finally, we invite medical experts to evaluate the clinical usefulness of LLMs | [
"Liu, Fenglin",
"Li, Zheng",
"Zhou, Hongjian",
"Yin, Qingyu",
"Yang, Jingfeng",
"Tang, Xianfeng",
"Luo, Chen",
"Zeng, Ming",
"Jiang, Haoming",
"Gao, Yifan",
"Nigam, Priyanka",
"Nag, Sreyashi",
"Yin, Bing",
"Hua, Yining",
"Zhou, Xuan",
"Rohanian, Omid",
"Thakur, Anshul",
"Clifton, Lei",
"Clifton, David A."
] | Large Language Models Are Poor Clinical Decision-Makers: A Comprehensive Benchmark | emnlp-main.759 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.760.bib | https://aclanthology.org/2024.emnlp-main.760/ | @inproceedings{zhang-etal-2024-holistic,
title = "Holistic Automated Red Teaming for Large Language Models through Top-Down Test Case Generation and Multi-turn Interaction",
author = "Zhang, Jinchuan and
Zhou, Yan and
Liu, Yaxin and
Li, Ziming and
Hu, Songlin",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.760",
pages = "13711--13736",
abstract = "Automated red teaming is an effective method for identifying misaligned behaviors in large language models (LLMs). Existing approaches, however, often focus primarily on improving attack success rates while overlooking the need for comprehensive test case coverage. Additionally, most of these methods are limited to single-turn red teaming, failing to capture the multi-turn dynamics of real-world human-machine interactions. To overcome these limitations, we propose **HARM** (**H**olistic **A**utomated **R**ed tea**M**ing), which scales up the diversity of test cases using a top-down approach based on an extensible, fine-grained risk taxonomy. Our method also leverages a novel fine-tuning strategy and reinforcement learning techniques to facilitate multi-turn adversarial probing in a human-like manner. Experimental results demonstrate that our framework enables a more systematic understanding of model vulnerabilities and offers more targeted guidance for the alignment process.",
}
| Automated red teaming is an effective method for identifying misaligned behaviors in large language models (LLMs). Existing approaches, however, often focus primarily on improving attack success rates while overlooking the need for comprehensive test case coverage. Additionally, most of these methods are limited to single-turn red teaming, failing to capture the multi-turn dynamics of real-world human-machine interactions. To overcome these limitations, we propose **HARM** (**H**olistic **A**utomated **R**ed tea**M**ing), which scales up the diversity of test cases using a top-down approach based on an extensible, fine-grained risk taxonomy. Our method also leverages a novel fine-tuning strategy and reinforcement learning techniques to facilitate multi-turn adversarial probing in a human-like manner. Experimental results demonstrate that our framework enables a more systematic understanding of model vulnerabilities and offers more targeted guidance for the alignment process. | [
"Zhang, Jinchuan",
"Zhou, Yan",
"Liu, Yaxin",
"Li, Ziming",
"Hu, Songlin"
] | Holistic Automated Red Teaming for Large Language Models through Top-Down Test Case Generation and Multi-turn Interaction | emnlp-main.760 | Poster | 2409.16783 | [
"https://github.com/jc-ryan/holistic_automated_red_teaming"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.761.bib | https://aclanthology.org/2024.emnlp-main.761/ | @inproceedings{pham-nguyen-2024-householder,
title = "Householder Pseudo-Rotation: A Novel Approach to Activation Editing in {LLM}s with Direction-Magnitude Perspective",
author = "Pham, Van-Cuong and
Nguyen, Thien Huu",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.761",
pages = "13737--13751",
abstract = "Activation Editing, which involves directly editting the internal representations of large language models (LLMs) to alter their behavior and achieve desired properties, has emerged as a promising area of research. Existing works primarily treat LLMs{'} activations as points in space and modify them by adding steering vectors. We show that doing so would break the magnitude consistency of the activation vectors in LLMs. To overcome this shortcoming, we propose a novel editing method that views activations in terms of their directions and magnitudes. Our method, which we name Householder Pseudo-Rotation (HPR), mimics the rotation transformation, thus preserving activation norm and resulting in an improved performance on various safety benchmarks.",
}
| Activation Editing, which involves directly editting the internal representations of large language models (LLMs) to alter their behavior and achieve desired properties, has emerged as a promising area of research. Existing works primarily treat LLMs{'} activations as points in space and modify them by adding steering vectors. We show that doing so would break the magnitude consistency of the activation vectors in LLMs. To overcome this shortcoming, we propose a novel editing method that views activations in terms of their directions and magnitudes. Our method, which we name Householder Pseudo-Rotation (HPR), mimics the rotation transformation, thus preserving activation norm and resulting in an improved performance on various safety benchmarks. | [
"Pham, Van-Cuong",
"Nguyen, Thien Huu"
] | Householder Pseudo-Rotation: A Novel Approach to Activation Editing in LLMs with Direction-Magnitude Perspective | emnlp-main.761 | Poster | 2409.10053 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.762.bib | https://aclanthology.org/2024.emnlp-main.762/ | @inproceedings{kim-etal-2024-dynamicer,
title = "{D}ynamic{ER}: Resolving Emerging Mentions to Dynamic Entities for {RAG}",
author = "Kim, Jinyoung and
Ko, Dayoon and
Kim, Gunhee",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.762",
pages = "13752--13770",
abstract = "In the rapidly evolving landscape of language, resolving new linguistic expressions in continuously updating knowledge bases remains a formidable challenge. This challenge becomes critical in retrieval-augmented generation (RAG) with knowledge bases, as emerging expressions hinder the retrieval of relevant documents, leading to generator hallucinations. To address this issue, we introduce a novel task aimed at resolving emerging mentions to dynamic entities and present DynamicER benchmark. Our benchmark includes dynamic entity mention resolution and entity-centric knowledge-intensive QA task, evaluating entity linking and RAG model{'}s adaptability to new expressions, respectively. We discovered that current entity linking models struggle to link these new expressions to entities. Therefore, we propose a temporal segmented clustering method with continual adaptation, effectively managing the temporal dynamics of evolving entities and emerging mentions. Extensive experiments demonstrate that our method outperforms existing baselines, enhancing RAG model performance on QA task with resolved mentions.",
}
| In the rapidly evolving landscape of language, resolving new linguistic expressions in continuously updating knowledge bases remains a formidable challenge. This challenge becomes critical in retrieval-augmented generation (RAG) with knowledge bases, as emerging expressions hinder the retrieval of relevant documents, leading to generator hallucinations. To address this issue, we introduce a novel task aimed at resolving emerging mentions to dynamic entities and present DynamicER benchmark. Our benchmark includes dynamic entity mention resolution and entity-centric knowledge-intensive QA task, evaluating entity linking and RAG model{'}s adaptability to new expressions, respectively. We discovered that current entity linking models struggle to link these new expressions to entities. Therefore, we propose a temporal segmented clustering method with continual adaptation, effectively managing the temporal dynamics of evolving entities and emerging mentions. Extensive experiments demonstrate that our method outperforms existing baselines, enhancing RAG model performance on QA task with resolved mentions. | [
"Kim, Jinyoung",
"Ko, Dayoon",
"Kim, Gunhee"
] | DynamicER: Resolving Emerging Mentions to Dynamic Entities for RAG | emnlp-main.762 | Poster | 2410.11494 | [
"https://github.com/jiny1623/dynamicer"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.763.bib | https://aclanthology.org/2024.emnlp-main.763/ | @inproceedings{tran-etal-2024-preserving,
title = "Preserving Generalization of Language models in Few-shot Continual Relation Extraction",
author = "Tran, Quyen and
Thanh, Nguyen Xuan and
Anh, Nguyen Hoang and
Hai, Nam Le and
Le, Trung and
Ngo, Linh Van and
Nguyen, Thien Huu",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.763",
pages = "13771--13784",
abstract = "Few-shot Continual Relations Extraction (FCRE) is an emerging and dynamic area of study where models can sequentially integrate knowledge from new relations with limited labeled data while circumventing catastrophic forgetting and preserving prior knowledge from pre-trained backbones. In this work, we introduce a novel method that leverages often-discarded language model heads. By employing these components via a mutual information maximization strategy, our approach helps maintain prior knowledge from the pre-trained backbone and strategically aligns the primary classification head, thereby enhancing model performance. Furthermore, we explore the potential of Large Language Models (LLMs), renowned for their wealth of knowledge, in addressing FCRE challenges. Our comprehensive experimental results underscore the efficacy of the proposed method and offer valuable insights for future work.",
}
| Few-shot Continual Relations Extraction (FCRE) is an emerging and dynamic area of study where models can sequentially integrate knowledge from new relations with limited labeled data while circumventing catastrophic forgetting and preserving prior knowledge from pre-trained backbones. In this work, we introduce a novel method that leverages often-discarded language model heads. By employing these components via a mutual information maximization strategy, our approach helps maintain prior knowledge from the pre-trained backbone and strategically aligns the primary classification head, thereby enhancing model performance. Furthermore, we explore the potential of Large Language Models (LLMs), renowned for their wealth of knowledge, in addressing FCRE challenges. Our comprehensive experimental results underscore the efficacy of the proposed method and offer valuable insights for future work. | [
"Tran, Quyen",
"Thanh, Nguyen Xuan",
"Anh, Nguyen Hoang",
"Hai, Nam Le",
"Le, Trung",
"Ngo, Linh Van",
"Nguyen, Thien Huu"
] | Preserving Generalization of Language models in Few-shot Continual Relation Extraction | emnlp-main.763 | Poster | 2410.00334 | [
"https://github.com/thanhnx12/cre-via-mmi"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.764.bib | https://aclanthology.org/2024.emnlp-main.764/ | @inproceedings{laskar-etal-2024-systematic,
title = "A Systematic Survey and Critical Review on Evaluating Large Language Models: Challenges, Limitations, and Recommendations",
author = "Laskar, Md Tahmid Rahman and
Alqahtani, Sawsan and
Bari, M Saiful and
Rahman, Mizanur and
Khan, Mohammad Abdullah Matin and
Khan, Haidar and
Jahan, Israt and
Bhuiyan, Amran and
Tan, Chee Wei and
Parvez, Md Rizwan and
Hoque, Enamul and
Joty, Shafiq and
Huang, Jimmy",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.764",
pages = "13785--13816",
abstract = "Large Language Models (LLMs) have recently gained significant attention due to their remarkable capabilities in performing diverse tasks across various domains. However, a thorough evaluation of these models is crucial before deploying them in real-world applications to ensure they produce reliable performance. Despite the well-established importance of evaluating LLMs in the community, the complexity of the evaluation process has led to varied evaluation setups, causing inconsistencies in findings and interpretations. To address this, we systematically review the primary challenges and limitations causing these inconsistencies and unreliable evaluations in various steps of LLM evaluation. Based on our critical review, we present our perspectives and recommendations to ensure LLM evaluations are reproducible, reliable, and robust.",
}
| Large Language Models (LLMs) have recently gained significant attention due to their remarkable capabilities in performing diverse tasks across various domains. However, a thorough evaluation of these models is crucial before deploying them in real-world applications to ensure they produce reliable performance. Despite the well-established importance of evaluating LLMs in the community, the complexity of the evaluation process has led to varied evaluation setups, causing inconsistencies in findings and interpretations. To address this, we systematically review the primary challenges and limitations causing these inconsistencies and unreliable evaluations in various steps of LLM evaluation. Based on our critical review, we present our perspectives and recommendations to ensure LLM evaluations are reproducible, reliable, and robust. | [
"Laskar, Md Tahmid Rahman",
"Alqahtani, Sawsan",
"Bari, M Saiful",
"Rahman, Mizanur",
"Khan, Mohammad Abdullah Matin",
"Khan, Haidar",
"Jahan, Israt",
"Bhuiyan, Amran",
"Tan, Chee Wei",
"Parvez, Md Rizwan",
"Hoque, Enamul",
"Joty, Shafiq",
"Huang, Jimmy"
] | A Systematic Survey and Critical Review on Evaluating Large Language Models: Challenges, Limitations, and Recommendations | emnlp-main.764 | Oral | 2407.04069 | [
""
] | https://huggingface.co/papers/2407.04069 | 0 | 0 | 0 | 13 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.765.bib | https://aclanthology.org/2024.emnlp-main.765/ | @inproceedings{li-etal-2024-consecutive,
title = "Consecutive Batch Model Editing with {H}oo{K} Layers",
author = "Li, Shuaiyi and
Deng, Yang and
Cai, Deng and
Lu, Hongyuan and
Chen, Liang and
Lam, Wai",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.765",
pages = "13817--13833",
abstract = "As the typical retraining paradigm is unacceptably time- and resource-consuming, researchers are turning to model editing to find an effective way that supports both consecutive and batch scenarios to edit the model behavior directly. Despite all these practical expectations, existing model editing methods fail to realize all of them. Furthermore, the memory demands for such sequential model editing approaches tend to be prohibitive, frequently necessitating an external memory that grows incrementally over time. To cope with these challenges, we propose CoachHooK, a model editing method that simultaneously supports sequential and batch editing. CoachHooK is memory-friendly as it only needs a small amount of it to store several hook layers whose size remains unchanged over time. Experimental results demonstrate the superiority of our method over other batch-supportive model editing methods under both single-round and consecutive batch editing scenarios. Extensive analyses of CoachHooK have been conducted to verify the stability of our method over a number of consecutive steps.",
}
| As the typical retraining paradigm is unacceptably time- and resource-consuming, researchers are turning to model editing to find an effective way that supports both consecutive and batch scenarios to edit the model behavior directly. Despite all these practical expectations, existing model editing methods fail to realize all of them. Furthermore, the memory demands for such sequential model editing approaches tend to be prohibitive, frequently necessitating an external memory that grows incrementally over time. To cope with these challenges, we propose CoachHooK, a model editing method that simultaneously supports sequential and batch editing. CoachHooK is memory-friendly as it only needs a small amount of it to store several hook layers whose size remains unchanged over time. Experimental results demonstrate the superiority of our method over other batch-supportive model editing methods under both single-round and consecutive batch editing scenarios. Extensive analyses of CoachHooK have been conducted to verify the stability of our method over a number of consecutive steps. | [
"Li, Shuaiyi",
"Deng, Yang",
"Cai, Deng",
"Lu, Hongyuan",
"Chen, Liang",
"Lam, Wai"
] | Consecutive Batch Model Editing with HooK Layers | emnlp-main.765 | Poster | 2403.05330 | [
"https://github.com/syon-li/coachhook"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.766.bib | https://aclanthology.org/2024.emnlp-main.766/ | @inproceedings{ding-etal-2024-topic,
title = "Topic-Oriented Open Relation Extraction with A Priori Seed Generation",
author = "Ding, Linyi and
Xiao, Jinfeng and
Zhou, Sizhe and
Yang, Chaoqi and
Han, Jiawei",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.766",
pages = "13834--13845",
abstract = "The field of open relation extraction (ORE) has recently observed significant advancement thanks to the growing capability of large language models (LLMs). Nevertheless, challenges persist when ORE is performed on specific topics. Existing methods give sub-optimal results in five dimensions: factualness, topic relevance, informativeness, coverage, and uniformity. To improve topic-oriented ORE, we propose a zero-shot approach called PriORE: Open Relation Extraction with a Priori seed generation. PriORE leverages the built-in knowledge of LLMs to maintain a dynamic seed relation dictionary for the topic. The dictionary is initialized by seed relations generated from topic-relevant entity types and expanded during contextualized ORE. PriORE then reduces the randomness in generative ORE by converting it to a more robust relation classification task. Experiments show the approach empowers better topic-oriented control over the generated relations and thus improves ORE performance along the five dimensions, especially on specialized and narrow topics.",
}
| The field of open relation extraction (ORE) has recently observed significant advancement thanks to the growing capability of large language models (LLMs). Nevertheless, challenges persist when ORE is performed on specific topics. Existing methods give sub-optimal results in five dimensions: factualness, topic relevance, informativeness, coverage, and uniformity. To improve topic-oriented ORE, we propose a zero-shot approach called PriORE: Open Relation Extraction with a Priori seed generation. PriORE leverages the built-in knowledge of LLMs to maintain a dynamic seed relation dictionary for the topic. The dictionary is initialized by seed relations generated from topic-relevant entity types and expanded during contextualized ORE. PriORE then reduces the randomness in generative ORE by converting it to a more robust relation classification task. Experiments show the approach empowers better topic-oriented control over the generated relations and thus improves ORE performance along the five dimensions, especially on specialized and narrow topics. | [
"Ding, Linyi",
"Xiao, Jinfeng",
"Zhou, Sizhe",
"Yang, Chaoqi",
"Han, Jiawei"
] | Topic-Oriented Open Relation Extraction with A Priori Seed Generation | emnlp-main.766 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.767.bib | https://aclanthology.org/2024.emnlp-main.767/ | @inproceedings{li-ouyang-2024-related,
title = "Related Work and Citation Text Generation: A Survey",
author = "Li, Xiangci and
Ouyang, Jessica",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.767",
pages = "13846--13864",
abstract = "To convince readers of the novelty of their research paper, authors must perform a literature review and compose a coherent story that connects and relates prior works to the current work. This challenging nature of literature review writing makes automatic related work generation (RWG) academically and computationally interesting, and also makes it an excellent test bed for examining the capability of SOTA natural language processing (NLP) models. Since the initial proposal of the RWG task, its popularity has waxed and waned, following the capabilities of mainstream NLP approaches. In this work, we survey the zoo of RWG historical works, summarizing the key approaches and task definitions and discussing the ongoing challenges of RWG.",
}
| To convince readers of the novelty of their research paper, authors must perform a literature review and compose a coherent story that connects and relates prior works to the current work. This challenging nature of literature review writing makes automatic related work generation (RWG) academically and computationally interesting, and also makes it an excellent test bed for examining the capability of SOTA natural language processing (NLP) models. Since the initial proposal of the RWG task, its popularity has waxed and waned, following the capabilities of mainstream NLP approaches. In this work, we survey the zoo of RWG historical works, summarizing the key approaches and task definitions and discussing the ongoing challenges of RWG. | [
"Li, Xiangci",
"Ouyang, Jessica"
] | Related Work and Citation Text Generation: A Survey | emnlp-main.767 | Poster | 2404.11588 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.768.bib | https://aclanthology.org/2024.emnlp-main.768/ | @inproceedings{liu-etal-2024-curriculum,
title = "Curriculum Consistency Learning for Conditional Sentence Generation",
author = "Liu, Liangxin and
Liu, Xuebo and
Lian, Lian and
Cheng, Shengjun and
Rao, Jun and
Yu, Tengfei and
Deng, Hexuan and
Zhang, Min",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.768",
pages = "13865--13881",
abstract = "Consistency learning (CL) has proven to be a valuable technique for improving the robustness of models in conditional sentence generation (CSG) tasks by ensuring stable predictions across various input data forms. However, models augmented with CL often face challenges in optimizing consistency features, which can detract from their efficiency and effectiveness. To address these challenges, we introduce Curriculum Consistency Learning (CCL), a novel strategy that guides models to learn consistency in alignment with their current capacity to differentiate between features. CCL is designed around the inherent aspects of CL-related losses, promoting task independence and simplifying implementation. Implemented across four representative CSG tasks, including instruction tuning (IT) for large language models and machine translation (MT) in three modalities (text, speech, and vision), CCL demonstrates marked improvements. Specifically, it delivers +2.0 average accuracy point improvement compared with vanilla IT and an average increase of +0.7 in COMET scores over traditional CL methods in MT tasks. Our comprehensive analysis further indicates that models utilizing CCL are particularly adept at managing complex instances, showcasing the effectiveness and efficiency of CCL in improving CSG models. Code and scripts are available at https://github.com/xinxinxing/Curriculum-Consistency-Learning.",
}
| Consistency learning (CL) has proven to be a valuable technique for improving the robustness of models in conditional sentence generation (CSG) tasks by ensuring stable predictions across various input data forms. However, models augmented with CL often face challenges in optimizing consistency features, which can detract from their efficiency and effectiveness. To address these challenges, we introduce Curriculum Consistency Learning (CCL), a novel strategy that guides models to learn consistency in alignment with their current capacity to differentiate between features. CCL is designed around the inherent aspects of CL-related losses, promoting task independence and simplifying implementation. Implemented across four representative CSG tasks, including instruction tuning (IT) for large language models and machine translation (MT) in three modalities (text, speech, and vision), CCL demonstrates marked improvements. Specifically, it delivers +2.0 average accuracy point improvement compared with vanilla IT and an average increase of +0.7 in COMET scores over traditional CL methods in MT tasks. Our comprehensive analysis further indicates that models utilizing CCL are particularly adept at managing complex instances, showcasing the effectiveness and efficiency of CCL in improving CSG models. Code and scripts are available at https://github.com/xinxinxing/Curriculum-Consistency-Learning. | [
"Liu, Liangxin",
"Liu, Xuebo",
"Lian, Lian",
"Cheng, Shengjun",
"Rao, Jun",
"Yu, Tengfei",
"Deng, Hexuan",
"Zhang, Min"
] | Curriculum Consistency Learning for Conditional Sentence Generation | emnlp-main.768 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.769.bib | https://aclanthology.org/2024.emnlp-main.769/ | @inproceedings{bertolazzi-etal-2024-systematic,
title = "A Systematic Analysis of Large Language Models as Soft Reasoners: The Case of Syllogistic Inferences",
author = "Bertolazzi, Leonardo and
Gatt, Albert and
Bernardi, Raffaella",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.769",
pages = "13882--13905",
abstract = "The reasoning abilities of Large Language Models (LLMs) are becoming a central focus of study in NLP. In this paper, we consider the case of syllogistic reasoning, an area of deductive reasoning studied extensively in logic and cognitive psychology. Previous research has shown that pre-trained LLMs exhibit reasoning biases, such as content effects, avoid answering that no conclusion follows, align with human difficulties, and struggle with multi-step reasoning. We contribute to this research line by systematically investigating the effects of chain-of-thought reasoning, in-context learning (ICL), and supervised fine-tuning (SFT) on syllogistic reasoning, considering syllogisms with conclusions that support or violate world knowledge and with multiple premises. Crucially, we go beyond the standard focus on accuracy, with an in-depth analysis of the conclusions generated by the models. Our results suggest that the behavior of pre-trained LLMs can be explained by heuristics studied in cognitive science and that both ICL and SFT improve model performance on valid inferences, although only the latter can mitigate most reasoning biases while being consistent.",
}
| The reasoning abilities of Large Language Models (LLMs) are becoming a central focus of study in NLP. In this paper, we consider the case of syllogistic reasoning, an area of deductive reasoning studied extensively in logic and cognitive psychology. Previous research has shown that pre-trained LLMs exhibit reasoning biases, such as content effects, avoid answering that no conclusion follows, align with human difficulties, and struggle with multi-step reasoning. We contribute to this research line by systematically investigating the effects of chain-of-thought reasoning, in-context learning (ICL), and supervised fine-tuning (SFT) on syllogistic reasoning, considering syllogisms with conclusions that support or violate world knowledge and with multiple premises. Crucially, we go beyond the standard focus on accuracy, with an in-depth analysis of the conclusions generated by the models. Our results suggest that the behavior of pre-trained LLMs can be explained by heuristics studied in cognitive science and that both ICL and SFT improve model performance on valid inferences, although only the latter can mitigate most reasoning biases while being consistent. | [
"Bertolazzi, Leonardo",
"Gatt, Albert",
"Bernardi, Raffaella"
] | A Systematic Analysis of Large Language Models as Soft Reasoners: The Case of Syllogistic Inferences | emnlp-main.769 | Poster | 2406.11341 | [
"https://github.com/leobertolazzi/soft-syllogistic-reasoners"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.770.bib | https://aclanthology.org/2024.emnlp-main.770/ | @inproceedings{jiang-etal-2024-pre,
title = "Pre-training Cross-lingual Open Domain Question Answering with Large-scale Synthetic Supervision",
author = "Jiang, Fan and
Drummond, Tom and
Cohn, Trevor",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.770",
pages = "13906--13933",
}
| No abstract found | [
"Jiang, Fan",
"Drummond, Tom",
"Cohn, Trevor"
] | Pre-training Cross-lingual Open Domain Question Answering with Large-scale Synthetic Supervision | emnlp-main.770 | Poster | 2402.16508 | [
"https://github.com/fantabulous-j/class"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.771.bib | https://aclanthology.org/2024.emnlp-main.771/ | @inproceedings{gaido-etal-2024-mosel,
title = "{MOSEL}: 950,000 Hours of Speech Data for Open-Source Speech Foundation Model Training on {EU} Languages",
author = "Gaido, Marco and
Papi, Sara and
Bentivogli, Luisa and
Brutti, Alessio and
Cettolo, Mauro and
Gretter, Roberto and
Matassoni, Marco and
Nabih, Mohamed and
Negri, Matteo",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.771",
pages = "13934--13947",
abstract = "The rise of foundation models (FMs), coupled with regulatory efforts addressing their risks and impacts, has sparked significant interest in open-source models. However, existing speech FMs (SFMs) fall short of full compliance with the open-source principles, even if claimed otherwise, as no existing SFM has model weights, code, and training data publicly available under open-source terms. In this work, we take the first step toward filling this gap by focusing on the 24 official languages of the European Union (EU). We collect suitable training data by surveying automatic speech recognition datasets and unlabeled speech corpora under open-source compliant licenses, for a total of 950k hours. Additionally, we release automatic transcripts for 441k hours of unlabeled data under the permissive CC-BY license, thereby facilitating the creation of open-source SFMs for the EU languages.",
}
| The rise of foundation models (FMs), coupled with regulatory efforts addressing their risks and impacts, has sparked significant interest in open-source models. However, existing speech FMs (SFMs) fall short of full compliance with the open-source principles, even if claimed otherwise, as no existing SFM has model weights, code, and training data publicly available under open-source terms. In this work, we take the first step toward filling this gap by focusing on the 24 official languages of the European Union (EU). We collect suitable training data by surveying automatic speech recognition datasets and unlabeled speech corpora under open-source compliant licenses, for a total of 950k hours. Additionally, we release automatic transcripts for 441k hours of unlabeled data under the permissive CC-BY license, thereby facilitating the creation of open-source SFMs for the EU languages. | [
"Gaido, Marco",
"Papi, Sara",
"Bentivogli, Luisa",
"Brutti, Alessio",
"Cettolo, Mauro",
"Gretter, Roberto",
"Matassoni, Marco",
"Nabih, Mohamed",
"Negri, Matteo"
] | MOSEL: 950,000 Hours of Speech Data for Open-Source Speech Foundation Model Training on EU Languages | emnlp-main.771 | Poster | 2410.01036 | [
"https://github.com/hlt-mt/mosel"
] | https://huggingface.co/papers/2410.01036 | 3 | 14 | 2 | 9 | [] | [
"FBK-MT/mosel"
] | [
"sakaltcommunity/Grape"
] | [] | [
"FBK-MT/mosel"
] | [
"sakaltcommunity/Grape"
] | 1 |
https://aclanthology.org/2024.emnlp-main.772.bib | https://aclanthology.org/2024.emnlp-main.772/ | @inproceedings{lin-etal-2024-improving,
title = "Improving Knowledge Graph Completion with Structure-Aware Supervised Contrastive Learning",
author = "Lin, Jiashi and
Wang, Lifang and
Lu, Xinyu and
Hu, Zhongtian and
Zhang, Wei and
Lu, Wenxuan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.772",
pages = "13948--13959",
abstract = "Knowledge Graphs (KGs) often suffer from incomplete knowledge, which which restricts their utility. Recently, Contrastive Learning (CL) has been introduced to Knowledge Graph Completion (KGC), significantly improving the discriminative capabilities of KGC models and setting new benchmarks in performance. However, existing contrastive methods primarily focus on individual triples, overlooking the broader structural connectivities and topologies of KGs. This narrow focus limits a comprehensive understanding of the graph{'}s structural knowledge. To address this gap, we propose StructKGC, a novel contrastive learning framework designed to flexibly accommodate the diverse topologies inherent in KGs. Additionally, we introduce four contrastive tasks specifically tailored to KG data: Vertex-level CL, Neighbor-level CL, Path-level CL, and Relation composition level CL. These tasks are trained synergistically during the fine-tuning of pre-trained language models (PLMs), allowing for a more nuanced capture of subgraph semantics. To validate the effectiveness of our method, we perform a comprehensive set of experiments on several real-world datasets. The experimental results demonstrate that our approach achieves SOTA performance under standard supervised and low-resource settings. Furthermore, the different levels of structure-aware tasks introduced can mutually reinforce each other, leading to consistent performance improvements.",
}
| Knowledge Graphs (KGs) often suffer from incomplete knowledge, which which restricts their utility. Recently, Contrastive Learning (CL) has been introduced to Knowledge Graph Completion (KGC), significantly improving the discriminative capabilities of KGC models and setting new benchmarks in performance. However, existing contrastive methods primarily focus on individual triples, overlooking the broader structural connectivities and topologies of KGs. This narrow focus limits a comprehensive understanding of the graph{'}s structural knowledge. To address this gap, we propose StructKGC, a novel contrastive learning framework designed to flexibly accommodate the diverse topologies inherent in KGs. Additionally, we introduce four contrastive tasks specifically tailored to KG data: Vertex-level CL, Neighbor-level CL, Path-level CL, and Relation composition level CL. These tasks are trained synergistically during the fine-tuning of pre-trained language models (PLMs), allowing for a more nuanced capture of subgraph semantics. To validate the effectiveness of our method, we perform a comprehensive set of experiments on several real-world datasets. The experimental results demonstrate that our approach achieves SOTA performance under standard supervised and low-resource settings. Furthermore, the different levels of structure-aware tasks introduced can mutually reinforce each other, leading to consistent performance improvements. | [
"Lin, Jiashi",
"Wang, Lifang",
"Lu, Xinyu",
"Hu, Zhongtian",
"Zhang, Wei",
"Lu, Wenxuan"
] | Improving Knowledge Graph Completion with Structure-Aware Supervised Contrastive Learning | emnlp-main.772 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.773.bib | https://aclanthology.org/2024.emnlp-main.773/ | @inproceedings{basirat-hemmati-2024-contribution,
title = "Contribution of Linguistic Typology to {U}niversal {D}ependency Parsing: An Empirical Investigation",
author = "Basirat, Ali and
Hemmati, Navid Baradaran",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.773",
pages = "13960--13971",
abstract = "Universal Dependencies (UD) is a global initiative to create a standard annotation for the dependency syntax of human languages. Addressing its deviation from typological principles, this study presents an empirical investigation of a typologically motivated transformation of UD proposed by William Croft. Our findings underscore the significance of the transformations across diverse languages and highlight their advantages and limitations.",
}
| Universal Dependencies (UD) is a global initiative to create a standard annotation for the dependency syntax of human languages. Addressing its deviation from typological principles, this study presents an empirical investigation of a typologically motivated transformation of UD proposed by William Croft. Our findings underscore the significance of the transformations across diverse languages and highlight their advantages and limitations. | [
"Basirat, Ali",
"Hemmati, Navid Baradaran"
] | Contribution of Linguistic Typology to Universal Dependency Parsing: An Empirical Investigation | emnlp-main.773 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.774.bib | https://aclanthology.org/2024.emnlp-main.774/ | @inproceedings{periti-etal-2024-trotr,
title = "{TR}o{TR}: A Framework for Evaluating the Re-contextualization of Text Reuse",
author = "Periti, Francesco and
Cassotti, Pierluigi and
Montanelli, Stefano and
Tahmasebi, Nina and
Schlechtweg, Dominik",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.774",
pages = "13972--13990",
abstract = "Current approaches for detecting text reuse do not focus on recontextualization, i.e., how the new context(s) of a reused text differs from its original context(s). In this paper, we propose a novel framework called TRoTR that relies on the notion of topic relatedness for evaluating the diachronic change of context in which text is reused. TRoTR includes two NLP tasks: TRiC and TRaC. TRiC is designed to evaluate the topic relatedness between a pair of recontextualizations. TRaC is designed to evaluate the overall topic variation within a set of recontextualizations. We also provide a curated TRoTR benchmark of biblical text reuse, human-annotated with topic relatedness. The benchmark exhibits an inter-annotator agreement of .811. We evaluate multiple, established SBERT models on the TRoTR tasks and find that they exhibit greater sensitivity to textual similarity than topic relatedness. Our experiments show that fine-tuning these models can mitigate such a kind of sensitivity.",
}
| Current approaches for detecting text reuse do not focus on recontextualization, i.e., how the new context(s) of a reused text differs from its original context(s). In this paper, we propose a novel framework called TRoTR that relies on the notion of topic relatedness for evaluating the diachronic change of context in which text is reused. TRoTR includes two NLP tasks: TRiC and TRaC. TRiC is designed to evaluate the topic relatedness between a pair of recontextualizations. TRaC is designed to evaluate the overall topic variation within a set of recontextualizations. We also provide a curated TRoTR benchmark of biblical text reuse, human-annotated with topic relatedness. The benchmark exhibits an inter-annotator agreement of .811. We evaluate multiple, established SBERT models on the TRoTR tasks and find that they exhibit greater sensitivity to textual similarity than topic relatedness. Our experiments show that fine-tuning these models can mitigate such a kind of sensitivity. | [
"Periti, Francesco",
"Cassotti, Pierluigi",
"Montanelli, Stefano",
"Tahmasebi, Nina",
"Schlechtweg, Dominik"
] | TRoTR: A Framework for Evaluating the Re-contextualization of Text Reuse | emnlp-main.774 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.775.bib | https://aclanthology.org/2024.emnlp-main.775/ | @inproceedings{wei-etal-2024-structured,
title = "Structured Optimal Brain Pruning for Large Language Models",
author = "Wei, Jiateng and
Lu, Quan and
Jiang, Ning and
Li, Siqi and
Xiang, Jingyang and
Chen, Jun and
Liu, Yong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.775",
pages = "13991--14007",
abstract = "The massive parameters and computational demands hinder the widespread application of Large Language Models (LLMs). Network pruning provides a practical solution to this problem. However, existing pruning works for LLMs mainly focus on unstructured pruning or necessitate post-pruning fine-tuning. The former relies on special hardware to accelerate computation, while the latter may need substantial computational resources. In this paper, we introduce a retraining-free structured pruning method called SoBP (Structured Optimal Brain Pruning). It leverages global first-order information to select pruning structures, then refines them with a local greedy approach, and finally adopts module-wise reconstruction to mitigate information loss. We assess the effectiveness of SoBP across 14 models from 3 LLM families on 8 distinct datasets. Experimental results demonstrate that SoBP outperforms current state-of-the-art methods.",
}
| The massive parameters and computational demands hinder the widespread application of Large Language Models (LLMs). Network pruning provides a practical solution to this problem. However, existing pruning works for LLMs mainly focus on unstructured pruning or necessitate post-pruning fine-tuning. The former relies on special hardware to accelerate computation, while the latter may need substantial computational resources. In this paper, we introduce a retraining-free structured pruning method called SoBP (Structured Optimal Brain Pruning). It leverages global first-order information to select pruning structures, then refines them with a local greedy approach, and finally adopts module-wise reconstruction to mitigate information loss. We assess the effectiveness of SoBP across 14 models from 3 LLM families on 8 distinct datasets. Experimental results demonstrate that SoBP outperforms current state-of-the-art methods. | [
"Wei, Jiateng",
"Lu, Quan",
"Jiang, Ning",
"Li, Siqi",
"Xiang, Jingyang",
"Chen, Jun",
"Liu, Yong"
] | Structured Optimal Brain Pruning for Large Language Models | emnlp-main.775 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.776.bib | https://aclanthology.org/2024.emnlp-main.776/ | @inproceedings{periti-etal-2024-automatically,
title = "Automatically Generated Definitions and their utility for Modeling Word Meaning",
author = "Periti, Francesco and
Alfter, David and
Tahmasebi, Nina",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.776",
pages = "14008--14026",
abstract = "Modeling lexical semantics is a challenging task, often suffering from interpretability pitfalls. In this paper, we delve into the generation of dictionary-like sense definitions and explore their utility for modeling word meaning. We fine-tuned two Llama models and include an existing T5-based model in our evaluation. Firstly, we evaluate the quality of the generated definitions on existing English benchmarks, setting new state-of-the-art results for the Definition Generation task. Next, we explore the use of definitions generated by our models as intermediate representations subsequently encoded as sentence embeddings. We evaluate this approach on lexical semantics tasks such as the Word-in-Context, Word Sense Induction, and Lexical Semantic Change, setting new state-of-the-art results in all three tasks when compared to unsupervised baselines.",
}
| Modeling lexical semantics is a challenging task, often suffering from interpretability pitfalls. In this paper, we delve into the generation of dictionary-like sense definitions and explore their utility for modeling word meaning. We fine-tuned two Llama models and include an existing T5-based model in our evaluation. Firstly, we evaluate the quality of the generated definitions on existing English benchmarks, setting new state-of-the-art results for the Definition Generation task. Next, we explore the use of definitions generated by our models as intermediate representations subsequently encoded as sentence embeddings. We evaluate this approach on lexical semantics tasks such as the Word-in-Context, Word Sense Induction, and Lexical Semantic Change, setting new state-of-the-art results in all three tasks when compared to unsupervised baselines. | [
"Periti, Francesco",
"Alfter, David",
"Tahmasebi, Nina"
] | Automatically Generated Definitions and their utility for Modeling Word Meaning | emnlp-main.776 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.777.bib | https://aclanthology.org/2024.emnlp-main.777/ | @inproceedings{wang-etal-2024-code,
title = "How Do Your Code {LLM}s perform? Empowering Code Instruction Tuning with Really Good Data",
author = "Wang, Yejie and
He, Keqing and
Fu, Dayuan and
GongQue, Zhuoma and
Xu, Heyang and
Chen, Yanxu and
Wang, Zhexu and
Fu, Yujia and
Dong, Guanting and
Diao, Muxi and
Wang, Jingang and
Zhang, Mengdi and
Cai, Xunliang and
Xu, Weiran",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.777",
pages = "14027--14043",
abstract = "Recently, there has been a growing interest in studying how to construct better code instruction tuning data. However, we observe Code models trained with these datasets exhibit high performance on HumanEval but perform worse on other benchmarks such as LiveCodeBench. Upon further investigation, we find that many datasets suffer from severe data leakage. After cleaning up most of the leaked data, some well-known high-quality datasets perform poorly. This discovery reveals a new challenge: identifying which dataset genuinely qualify as high-quality code instruction data. To address this, we propose an efficient code data pruning strategy for selecting good samples. Our approach is based on three dimensions: instruction complexity, response quality, and instruction diversity. Based on our selected data, we present XCoder, a family of models finetuned from LLaMA3. Our experiments show Xcoder achieves new state-of-the-art performance using fewer training data, which verify the effectiveness of our data strategy. Moreover, we perform a comprehensive analysis on the data composition and find existing code datasets have different characteristics according to their construction methods, which provide new insights for future code LLMs.",
}
| Recently, there has been a growing interest in studying how to construct better code instruction tuning data. However, we observe Code models trained with these datasets exhibit high performance on HumanEval but perform worse on other benchmarks such as LiveCodeBench. Upon further investigation, we find that many datasets suffer from severe data leakage. After cleaning up most of the leaked data, some well-known high-quality datasets perform poorly. This discovery reveals a new challenge: identifying which dataset genuinely qualify as high-quality code instruction data. To address this, we propose an efficient code data pruning strategy for selecting good samples. Our approach is based on three dimensions: instruction complexity, response quality, and instruction diversity. Based on our selected data, we present XCoder, a family of models finetuned from LLaMA3. Our experiments show Xcoder achieves new state-of-the-art performance using fewer training data, which verify the effectiveness of our data strategy. Moreover, we perform a comprehensive analysis on the data composition and find existing code datasets have different characteristics according to their construction methods, which provide new insights for future code LLMs. | [
"Wang, Yejie",
"He, Keqing",
"Fu, Dayuan",
"GongQue, Zhuoma",
"Xu, Heyang",
"Chen, Yanxu",
"Wang, Zhexu",
"Fu, Yujia",
"Dong, Guanting",
"Diao, Muxi",
"Wang, Jingang",
"Zhang, Mengdi",
"Cai, Xunliang",
"Xu, Weiran"
] | How Do Your Code LLMs perform? Empowering Code Instruction Tuning with Really Good Data | emnlp-main.777 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.778.bib | https://aclanthology.org/2024.emnlp-main.778/ | @inproceedings{sun-etal-2024-mair,
title = "{MAIR}: A Massive Benchmark for Evaluating Instructed Retrieval",
author = "Sun, Weiwei and
Shi, Zhengliang and
Long, Wu Jiu and
Yan, Lingyong and
Ma, Xinyu and
Liu, Yiding and
Cao, Min and
Yin, Dawei and
Ren, Zhaochun",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.778",
pages = "14044--14067",
abstract = "Recent information retrieval (IR) models are pre-trained and instruction-tuned on massive datasets and tasks, enabling them to perform well on a wide range of tasks and potentially generalize to unseen tasks with instructions. However, existing IR benchmarks focus on a limited scope of tasks, making them insufficient for evaluating the latest IR models. In this paper, we propose MAIR (Massive Instructed Retrieval Benchmark), a heterogeneous IR benchmark that includes 126 distinct IR tasks across 6 domains, collected from existing datasets. We benchmark state-of-the-art instruction-tuned text embedding models and re-ranking models. Our experiments reveal that instruction-tuned models generally achieve superior performance compared to non-instruction-tuned models on MAIR Additionally, our results suggest that current instruction-tuned text embedding models and re-ranking models still lack effectiveness in specific long-tail tasks. MAIR is publicly available at https://github.com/sunnweiwei/Mair.",
}
| Recent information retrieval (IR) models are pre-trained and instruction-tuned on massive datasets and tasks, enabling them to perform well on a wide range of tasks and potentially generalize to unseen tasks with instructions. However, existing IR benchmarks focus on a limited scope of tasks, making them insufficient for evaluating the latest IR models. In this paper, we propose MAIR (Massive Instructed Retrieval Benchmark), a heterogeneous IR benchmark that includes 126 distinct IR tasks across 6 domains, collected from existing datasets. We benchmark state-of-the-art instruction-tuned text embedding models and re-ranking models. Our experiments reveal that instruction-tuned models generally achieve superior performance compared to non-instruction-tuned models on MAIR Additionally, our results suggest that current instruction-tuned text embedding models and re-ranking models still lack effectiveness in specific long-tail tasks. MAIR is publicly available at https://github.com/sunnweiwei/Mair. | [
"Sun, Weiwei",
"Shi, Zhengliang",
"Long, Wu Jiu",
"Yan, Lingyong",
"Ma, Xinyu",
"Liu, Yiding",
"Cao, Min",
"Yin, Dawei",
"Ren, Zhaochun"
] | MAIR: A Massive Benchmark for Evaluating Instructed Retrieval | emnlp-main.778 | Poster | 2410.10127 | [
"https://github.com/sunnweiwei/mair"
] | https://huggingface.co/papers/2410.10127 | 0 | 0 | 0 | 9 | [] | [
"MAIR-Bench/MAIR-Queries",
"MAIR-Bench/MAIR-Docs"
] | [] | [] | [
"MAIR-Bench/MAIR-Queries",
"MAIR-Bench/MAIR-Docs"
] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.779.bib | https://aclanthology.org/2024.emnlp-main.779/ | @inproceedings{yu-etal-2024-rethinking,
title = "Rethinking the Evaluation of In-Context Learning for {LLM}s",
author = "Yu, Guoxin and
Liu, Lemao and
Yu, Mo and
Yu, Yue and
Ao, Xiang",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.779",
pages = "14068--14082",
abstract = "In-context learning (ICL) has demonstrated excellent performance across various downstream NLP tasks, especially when synergized with powerful large language models (LLMs). Existing studies evaluate ICL methods primarily based on downstream task performance. This evaluation protocol overlooks the significant cost associated with the demonstration configuration process, i.e., tuning the demonstration as the ICL prompt. However, in this work, we point out that the evaluation protocol leads to unfair comparisons and potentially biased evaluation, because we surprisingly find the correlation between the configuration costs and task performance. Then we call for a two-dimensional evaluation paradigm that considers both of these aspects, facilitating a fairer comparison.Finally, based on our empirical finding that the optimized demonstration on one language model generalizes across language models of different sizes, we introduce a simple yet efficient strategy that can be applied to any ICL method as a plugin, yielding a better trade-off between the two dimensions according to the proposed evaluation paradigm.",
}
| In-context learning (ICL) has demonstrated excellent performance across various downstream NLP tasks, especially when synergized with powerful large language models (LLMs). Existing studies evaluate ICL methods primarily based on downstream task performance. This evaluation protocol overlooks the significant cost associated with the demonstration configuration process, i.e., tuning the demonstration as the ICL prompt. However, in this work, we point out that the evaluation protocol leads to unfair comparisons and potentially biased evaluation, because we surprisingly find the correlation between the configuration costs and task performance. Then we call for a two-dimensional evaluation paradigm that considers both of these aspects, facilitating a fairer comparison.Finally, based on our empirical finding that the optimized demonstration on one language model generalizes across language models of different sizes, we introduce a simple yet efficient strategy that can be applied to any ICL method as a plugin, yielding a better trade-off between the two dimensions according to the proposed evaluation paradigm. | [
"Yu, Guoxin",
"Liu, Lemao",
"Yu, Mo",
"Yu, Yue",
"Ao, Xiang"
] | Rethinking the Evaluation of In-Context Learning for LLMs | emnlp-main.779 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.780.bib | https://aclanthology.org/2024.emnlp-main.780/ | @inproceedings{laurito-etal-2024-cluster,
title = "Cluster-Norm for Unsupervised Probing of Knowledge",
author = {Laurito, Walter and
Maiya, Sharan and
Dhimo{\"\i}la, Gr{\'e}goire and
Yeung, Owen Ho Wan and
H{\"a}nni, Kaarel},
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.780",
pages = "14083--14112",
abstract = "The deployment of language models brings challenges in generating reliable text, especially when these models are fine-tuned with human preferences. To extract the encoded knowledge in these models without (potentially) biased human labels, unsupervised probing techniques like Contrast-Consistent Search (CCS) have been developed (Burns et al., 2022). However, salient but unrelated features in activation space can mislead these probes (Farquhar et al., 2023). Addressing this, we propose a cluster-normalization method to minimize the impact of such features by clustering and normalizing activations of contrast pairs before applying unsupervised probing techniques. While this approach does not address the issue of distinguishing between latent knowledge and that portrayed by a simulated agent{---}a major issue in the literature of eliciting latent knowledge (Paul Christiano and Xu, 2021){---}it still significantly improves the accuracy of probes in identifying the intended knowledge amidst distractions.",
}
| The deployment of language models brings challenges in generating reliable text, especially when these models are fine-tuned with human preferences. To extract the encoded knowledge in these models without (potentially) biased human labels, unsupervised probing techniques like Contrast-Consistent Search (CCS) have been developed (Burns et al., 2022). However, salient but unrelated features in activation space can mislead these probes (Farquhar et al., 2023). Addressing this, we propose a cluster-normalization method to minimize the impact of such features by clustering and normalizing activations of contrast pairs before applying unsupervised probing techniques. While this approach does not address the issue of distinguishing between latent knowledge and that portrayed by a simulated agent{---}a major issue in the literature of eliciting latent knowledge (Paul Christiano and Xu, 2021){---}it still significantly improves the accuracy of probes in identifying the intended knowledge amidst distractions. | [
"Laurito, Walter",
"Maiya, Sharan",
"Dhimo{\\\"\\i}la, Gr{\\'e}goire",
"Yeung, Owen Ho Wan",
"H{\\\"a}nni, Kaarel"
] | Cluster-Norm for Unsupervised Probing of Knowledge | emnlp-main.780 | Poster | 2407.18712 | [
"https://github.com/cadenza-labs/cluster-normalization"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.781.bib | https://aclanthology.org/2024.emnlp-main.781/ | @inproceedings{biran-etal-2024-hopping,
title = "Hopping Too Late: Exploring the Limitations of Large Language Models on Multi-Hop Queries",
author = "Biran, Eden and
Gottesman, Daniela and
Yang, Sohee and
Geva, Mor and
Globerson, Amir",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.781",
pages = "14113--14130",
abstract = "Large language models (LLMs) can solve complex multi-step problems, but little is known about how these computations are implemented internally. Motivated by this, we study how LLMs answer multi-hop queries such as {``}The spouse of the performer of Imagine is{''}. These queries require two information extraction steps: a latent one for resolving the first hop ({``}the performer of Imagine{''}) into the bridge entity (John Lennon), and another for resolving the second hop ({``}the spouse of John Lennon{''}) into the target entity (Yoko Ono). Understanding how the latent step is computed internally is key to understanding the overall computation. By carefully analyzing the internal computations of transformer-based LLMs, we discover that the bridge entity is resolved in the early layers of the model. Then, only after this resolution, the two-hop query is solved in the later layers. Because the second hop commences in later layers, there could be cases where these layers no longer encode the necessary knowledge for correctly predicting the answer. Motivated by this, we propose a novel {``}back-patching{''} analysis method whereby a hidden representation from a later layer is patched back to an earlier layer. We find that in up to 66{\%} of previously incorrect cases there exists a back-patch that results in the correct generation of the answer, showing that the later layers indeed sometimes lack the needed functionality. Overall our methods and findings open further opportunities for understanding and improving latent reasoning in transformer-based LLMs.",
}
| Large language models (LLMs) can solve complex multi-step problems, but little is known about how these computations are implemented internally. Motivated by this, we study how LLMs answer multi-hop queries such as {``}The spouse of the performer of Imagine is{''}. These queries require two information extraction steps: a latent one for resolving the first hop ({``}the performer of Imagine{''}) into the bridge entity (John Lennon), and another for resolving the second hop ({``}the spouse of John Lennon{''}) into the target entity (Yoko Ono). Understanding how the latent step is computed internally is key to understanding the overall computation. By carefully analyzing the internal computations of transformer-based LLMs, we discover that the bridge entity is resolved in the early layers of the model. Then, only after this resolution, the two-hop query is solved in the later layers. Because the second hop commences in later layers, there could be cases where these layers no longer encode the necessary knowledge for correctly predicting the answer. Motivated by this, we propose a novel {``}back-patching{''} analysis method whereby a hidden representation from a later layer is patched back to an earlier layer. We find that in up to 66{\%} of previously incorrect cases there exists a back-patch that results in the correct generation of the answer, showing that the later layers indeed sometimes lack the needed functionality. Overall our methods and findings open further opportunities for understanding and improving latent reasoning in transformer-based LLMs. | [
"Biran, Eden",
"Gottesman, Daniela",
"Yang, Sohee",
"Geva, Mor",
"Globerson, Amir"
] | Hopping Too Late: Exploring the Limitations of Large Language Models on Multi-Hop Queries | emnlp-main.781 | Poster | 2406.12775 | [
"https://github.com/edenbiran/HoppingTooLate"
] | https://huggingface.co/papers/2406.12775 | 1 | 0 | 0 | 5 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.782.bib | https://aclanthology.org/2024.emnlp-main.782/ | @inproceedings{wu-etal-2024-enhancing-training,
title = "Enhancing Training Data Attribution for Large Language Models with Fitting Error Consideration",
author = "Wu, Kangxi and
Pang, Liang and
Shen, Huawei and
Cheng, Xueqi",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.782",
pages = "14131--14143",
abstract = "The black-box nature of large language models (LLMs) poses challenges in interpreting results, impacting issues such as data intellectual property protection and hallucination tracing. Training data attribution (TDA) methods are considered effective solutions to address these challenges.Most recent TDA methods rely on influence functions, assuming the model achieves minimized empirical risk. However, achieving this criterion is difficult, and sourcing accuracy can be compromised by fitting errors during model training. In this paper, we introduce a novel TDA method called Debias and Denoise Attribution (DDA), which enhances influence functions by addressing fitting errors. Specifically, the debias strategy seeks to improve the performance of influence functions by eliminating the knowledge bias present in the base model before fine-tuning, while the denoise strategy aims to reduce discrepancies in influence scores arising from varying degrees of fitting during the training process through smoothing techniques.Experimental results demonstrate that our method significantly outperforms existing approaches, achieving an averaged AUC of 91.64{\%}. Moreover, DDA exhibits strong generality and scalability across various sources and different-scale models like LLaMA2, QWEN2, and Mistral.",
}
| The black-box nature of large language models (LLMs) poses challenges in interpreting results, impacting issues such as data intellectual property protection and hallucination tracing. Training data attribution (TDA) methods are considered effective solutions to address these challenges.Most recent TDA methods rely on influence functions, assuming the model achieves minimized empirical risk. However, achieving this criterion is difficult, and sourcing accuracy can be compromised by fitting errors during model training. In this paper, we introduce a novel TDA method called Debias and Denoise Attribution (DDA), which enhances influence functions by addressing fitting errors. Specifically, the debias strategy seeks to improve the performance of influence functions by eliminating the knowledge bias present in the base model before fine-tuning, while the denoise strategy aims to reduce discrepancies in influence scores arising from varying degrees of fitting during the training process through smoothing techniques.Experimental results demonstrate that our method significantly outperforms existing approaches, achieving an averaged AUC of 91.64{\%}. Moreover, DDA exhibits strong generality and scalability across various sources and different-scale models like LLaMA2, QWEN2, and Mistral. | [
"Wu, Kangxi",
"Pang, Liang",
"Shen, Huawei",
"Cheng, Xueqi"
] | Enhancing Training Data Attribution for Large Language Models with Fitting Error Consideration | emnlp-main.782 | Poster | 2410.01285 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.783.bib | https://aclanthology.org/2024.emnlp-main.783/ | @inproceedings{koo-etal-2024-large,
title = "Where am {I}? Large Language Models Wandering between Semantics and Structures in Long Contexts",
author = "Koo, Seonmin and
Kim, Jinsung and
Jang, YoungJoon and
Park, Chanjun and
Lim, Heuiseok",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.783",
pages = "14144--14160",
abstract = "As the utilization of Large Language Models (LLMs) becomes more widespread, there is a growing demand for their ability to handle more complex and longer external knowledge across various use cases. Most existing evaluations of the open-ended question answering (ODQA) task, which necessitates the use of external knowledge, focus solely on whether the model provides the correct answer. However, even when LLMs answer correctly, they often fail to provide an obvious source for their responses. Therefore, it is necessary to jointly evaluate and verify the correctness of the answers and the appropriateness of grounded evidence in complex external contexts. To address this issue, we examine the phenomenon of discrepancies in abilities across two distinct tasks{---}QA and evidence selection{---}when performed simultaneously, from the perspective of task alignment. To verify LLMs{'} task alignment, we introduce a verification framework and resources considering both semantic relevancy and structural diversity of the given long context knowledge. Through extensive experiments and detailed analysis, we provide insights into the task misalignment between QA and evidence selection. Our code and resources will be available upon acceptance.",
}
| As the utilization of Large Language Models (LLMs) becomes more widespread, there is a growing demand for their ability to handle more complex and longer external knowledge across various use cases. Most existing evaluations of the open-ended question answering (ODQA) task, which necessitates the use of external knowledge, focus solely on whether the model provides the correct answer. However, even when LLMs answer correctly, they often fail to provide an obvious source for their responses. Therefore, it is necessary to jointly evaluate and verify the correctness of the answers and the appropriateness of grounded evidence in complex external contexts. To address this issue, we examine the phenomenon of discrepancies in abilities across two distinct tasks{---}QA and evidence selection{---}when performed simultaneously, from the perspective of task alignment. To verify LLMs{'} task alignment, we introduce a verification framework and resources considering both semantic relevancy and structural diversity of the given long context knowledge. Through extensive experiments and detailed analysis, we provide insights into the task misalignment between QA and evidence selection. Our code and resources will be available upon acceptance. | [
"Koo, Seonmin",
"Kim, Jinsung",
"Jang, YoungJoon",
"Park, Chanjun",
"Lim, Heuiseok"
] | Where am I? Large Language Models Wandering between Semantics and Structures in Long Contexts | emnlp-main.783 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.784.bib | https://aclanthology.org/2024.emnlp-main.784/ | @inproceedings{shu-etal-2024-karl,
title = "{KARL}: Knowledge-Aware Retrieval and Representations aid Retention and Learning in Students",
author = "Shu, Matthew and
Balepur, Nishant and
Feng, Shi and
Boyd-Graber, Jordan Lee",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.784",
pages = "14161--14178",
abstract = "Flashcard schedulers rely on 1) *student models* to predict the flashcards a student knows; and 2) *teaching policies* to pick which cards to show next via these predictions.Prior student models, however, just use study data like the student{'}s past responses, ignoring the text on cards. We propose **content-aware scheduling**, the first schedulers exploiting flashcard content.To give the first evidence that such schedulers enhance student learning, we build KARL, a simple but effective content-aware student model employing deep knowledge tracing (DKT), retrieval, and BERT to predict student recall.We train KARL by collecting a new dataset of 123,143 study logs on diverse trivia questions.KARL bests existing student models in AUC and calibration error.To ensure our improved predictions lead to better student learning, we create a novel delta-based teaching policy to deploy KARL online.Based on 32 study paths from 27 users, KARL improves learning efficiency over SOTA, showing KARL{'}s strength and encouraging researchers to look beyond historical study data to fully capture student abilities.",
}
| Flashcard schedulers rely on 1) *student models* to predict the flashcards a student knows; and 2) *teaching policies* to pick which cards to show next via these predictions.Prior student models, however, just use study data like the student{'}s past responses, ignoring the text on cards. We propose **content-aware scheduling**, the first schedulers exploiting flashcard content.To give the first evidence that such schedulers enhance student learning, we build KARL, a simple but effective content-aware student model employing deep knowledge tracing (DKT), retrieval, and BERT to predict student recall.We train KARL by collecting a new dataset of 123,143 study logs on diverse trivia questions.KARL bests existing student models in AUC and calibration error.To ensure our improved predictions lead to better student learning, we create a novel delta-based teaching policy to deploy KARL online.Based on 32 study paths from 27 users, KARL improves learning efficiency over SOTA, showing KARL{'}s strength and encouraging researchers to look beyond historical study data to fully capture student abilities. | [
"Shu, Matthew",
"Balepur, Nishant",
"Feng, Shi",
"Boyd-Graber, Jordan Lee"
] | KARL: Knowledge-Aware Retrieval and Representations aid Retention and Learning in Students | emnlp-main.784 | Poster | 2402.12291 | [
""
] | https://huggingface.co/papers/2402.12291 | 0 | 0 | 0 | 4 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.785.bib | https://aclanthology.org/2024.emnlp-main.785/ | @inproceedings{xiao-etal-2024-large,
title = "Large Language Models Can Be Contextual Privacy Protection Learners",
author = "Xiao, Yijia and
Jin, Yiqiao and
Bai, Yushi and
Wu, Yue and
Yang, Xianjun and
Luo, Xiao and
Yu, Wenchao and
Zhao, Xujiang and
Liu, Yanchi and
Gu, Quanquan and
Chen, Haifeng and
Wang, Wei and
Cheng, Wei",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.785",
pages = "14179--14201",
abstract = "The proliferation of Large Language Models (LLMs) has driven considerable interest in fine-tuning them with domain-specific data to create specialized language models. Nevertheless, such domain-specific fine-tuning data often contains contextually sensitive personally identifiable information (PII). Direct fine-tuning LLMs on this data without privacy protection poses a risk of data leakage of sensitive PII during inference time. To address this challenge, we introduce Contextual Privacy Protection Language Models (CPPLM), a novel paradigm for fine-tuning LLMs that effectively injects domain-specific knowledge while safeguarding inference-time data privacy. Our work offers a theoretical analysis for model design and delves into various techniques such as corpus curation, penalty-based unlikelihood in training loss, and instruction-based tuning, etc. Extensive experiments across diverse datasets and scenarios demonstrate the effectiveness of our approaches. In particular, instruction tuning with both positive and negative examples, stands out as a promising method, effectively protecting private data while enhancing the model{'}s knowledge. Our work underscores the potential for Large Language Models as robust contextual privacy protection learners.",
}
| The proliferation of Large Language Models (LLMs) has driven considerable interest in fine-tuning them with domain-specific data to create specialized language models. Nevertheless, such domain-specific fine-tuning data often contains contextually sensitive personally identifiable information (PII). Direct fine-tuning LLMs on this data without privacy protection poses a risk of data leakage of sensitive PII during inference time. To address this challenge, we introduce Contextual Privacy Protection Language Models (CPPLM), a novel paradigm for fine-tuning LLMs that effectively injects domain-specific knowledge while safeguarding inference-time data privacy. Our work offers a theoretical analysis for model design and delves into various techniques such as corpus curation, penalty-based unlikelihood in training loss, and instruction-based tuning, etc. Extensive experiments across diverse datasets and scenarios demonstrate the effectiveness of our approaches. In particular, instruction tuning with both positive and negative examples, stands out as a promising method, effectively protecting private data while enhancing the model{'}s knowledge. Our work underscores the potential for Large Language Models as robust contextual privacy protection learners. | [
"Xiao, Yijia",
"Jin, Yiqiao",
"Bai, Yushi",
"Wu, Yue",
"Yang, Xianjun",
"Luo, Xiao",
"Yu, Wenchao",
"Zhao, Xujiang",
"Liu, Yanchi",
"Gu, Quanquan",
"Chen, Haifeng",
"Wang, Wei",
"Cheng, Wei"
] | Large Language Models Can Be Contextual Privacy Protection Learners | emnlp-main.785 | Poster | [
"https://github.com/yijia-xiao/pplm"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.786.bib | https://aclanthology.org/2024.emnlp-main.786/ | @inproceedings{balepur-etal-2024-smart,
title = "A {SMART} Mnemonic Sounds like {``}Glue Tonic{''}: Mixing {LLM}s with Student Feedback to Make Mnemonic Learning Stick",
author = "Balepur, Nishant and
Shu, Matthew and
Hoyle, Alexander and
Robey, Alison and
Feng, Shi and
Goldfarb-Tarrant, Seraphina and
Boyd-Graber, Jordan Lee",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.786",
pages = "14202--14225",
abstract = "Keyword mnemonics are memorable explanations that link new terms to simpler keywords.Prior work generates mnemonics for students, but they do not train models using mnemonics students prefer and aid learning.We build SMART, a mnemonic generator trained on feedback from real students learning new terms.To train SMART, we first fine-tune LLaMA-2 on a curated set of user-written mnemonics.We then use LLM alignment to enhance SMART: we deploy mnemonics generated by SMART in a flashcard app to find preferences on mnemonics students favor.We gather 2684 preferences from 45 students across two types: **expressed** (inferred from ratings) and **observed** (inferred from student learning), yielding three key findings.First, expressed and observed preferences disagree; what students *think* is helpful does not always capture what is *truly* helpful.Second, Bayesian models can synthesize complementary data from multiple preference types into a single effectiveness signal.SMART is tuned via Direct Preference Optimization on this signal, which resolves ties and missing labels in the typical method of pairwise comparisons, augmenting data for LLM output quality gains. Third, mnemonic experts assess SMART as matching GPT-4 at much lower deployment costs, showing the utility of capturing diverse student feedback to align LLMs in education.",
}
| Keyword mnemonics are memorable explanations that link new terms to simpler keywords.Prior work generates mnemonics for students, but they do not train models using mnemonics students prefer and aid learning.We build SMART, a mnemonic generator trained on feedback from real students learning new terms.To train SMART, we first fine-tune LLaMA-2 on a curated set of user-written mnemonics.We then use LLM alignment to enhance SMART: we deploy mnemonics generated by SMART in a flashcard app to find preferences on mnemonics students favor.We gather 2684 preferences from 45 students across two types: **expressed** (inferred from ratings) and **observed** (inferred from student learning), yielding three key findings.First, expressed and observed preferences disagree; what students *think* is helpful does not always capture what is *truly* helpful.Second, Bayesian models can synthesize complementary data from multiple preference types into a single effectiveness signal.SMART is tuned via Direct Preference Optimization on this signal, which resolves ties and missing labels in the typical method of pairwise comparisons, augmenting data for LLM output quality gains. Third, mnemonic experts assess SMART as matching GPT-4 at much lower deployment costs, showing the utility of capturing diverse student feedback to align LLMs in education. | [
"Balepur, Nishant",
"Shu, Matthew",
"Hoyle, Alex",
"er",
"Robey, Alison",
"Feng, Shi",
"Goldfarb-Tarrant, Seraphina",
"Boyd-Graber, Jordan Lee"
] | A SMART Mnemonic Sounds like “Glue Tonic”: Mixing LLMs with Student Feedback to Make Mnemonic Learning Stick | emnlp-main.786 | Poster | [
"https://github.com/nbalepur/mnemonic"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.787.bib | https://aclanthology.org/2024.emnlp-main.787/ | @inproceedings{wu-etal-2024-mixture-skills,
title = "Mixture-of-Skills: Learning to Optimize Data Usage for Fine-Tuning Large Language Models",
author = "Wu, Minghao and
Vu, Thuy-Trang and
Qu, Lizhen and
Haf, Reza",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.787",
pages = "14226--14240",
abstract = "Large language models (LLMs) are typically fine-tuned on diverse and extensive datasets sourced from various origins to develop a comprehensive range of skills, such as writing, reasoning, chatting, coding, and more. Each skill has unique characteristics, and these datasets are often heterogeneous and imbalanced, making the fine-tuning process highly challenging. Balancing the development of each skill while ensuring the model maintains its overall performance requires sophisticated techniques and careful dataset curation. In this work, we propose a general, model-agnostic, reinforcement learning framework, Mixture-of-Skills (MoS), that learns to optimize data usage automatically during the fine-tuning process. This framework ensures the optimal comprehensive skill development of LLMs by dynamically adjusting the focus on different datasets based on their current learning state. To validate the effectiveness of MoS, we conduct extensive experiments using three diverse LLM backbones on two widely used benchmarks and demonstrate that MoS substantially enhances model performance. Building on the success of MoS, we propose MoSpec, an adaptation for task-specific fine-tuning, which harnesses the utilities of various datasets for a specific purpose. Our work underlines the significance of dataset rebalancing and present MoS as a powerful, general solution for optimizing data usage in the fine-tuning of LLMs for various purposes.",
}
| Large language models (LLMs) are typically fine-tuned on diverse and extensive datasets sourced from various origins to develop a comprehensive range of skills, such as writing, reasoning, chatting, coding, and more. Each skill has unique characteristics, and these datasets are often heterogeneous and imbalanced, making the fine-tuning process highly challenging. Balancing the development of each skill while ensuring the model maintains its overall performance requires sophisticated techniques and careful dataset curation. In this work, we propose a general, model-agnostic, reinforcement learning framework, Mixture-of-Skills (MoS), that learns to optimize data usage automatically during the fine-tuning process. This framework ensures the optimal comprehensive skill development of LLMs by dynamically adjusting the focus on different datasets based on their current learning state. To validate the effectiveness of MoS, we conduct extensive experiments using three diverse LLM backbones on two widely used benchmarks and demonstrate that MoS substantially enhances model performance. Building on the success of MoS, we propose MoSpec, an adaptation for task-specific fine-tuning, which harnesses the utilities of various datasets for a specific purpose. Our work underlines the significance of dataset rebalancing and present MoS as a powerful, general solution for optimizing data usage in the fine-tuning of LLMs for various purposes. | [
"Wu, Minghao",
"Vu, Thuy-Trang",
"Qu, Lizhen",
"Haf, Reza"
] | Mixture-of-Skills: Learning to Optimize Data Usage for Fine-Tuning Large Language Models | emnlp-main.787 | Poster | 2406.08811 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.788.bib | https://aclanthology.org/2024.emnlp-main.788/ | @inproceedings{park-etal-2024-moltres,
title = "{M}ol{TRES}: Improving Chemical Language Representation Learning for Molecular Property Prediction",
author = "Park, Jun-Hyung and
Kim, Yeachan and
Lee, Mingyu and
Park, Hyuntae and
Lee, SangKeun",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.788",
pages = "14241--14254",
abstract = "Chemical representation learning has gained increasing interest due to the limited availability of supervised data in fields such as drug and materials design. This interest particularly extends to chemical language representation learning, which involves pre-training Transformers on SMILES sequences - textual descriptors of molecules. Despite its success in molecular property prediction, current practices often lead to overfitting and limited scalability due to early convergence. In this paper, we introduce a novel chemical language representation learning framework, called MolTRES, to address these issues. MolTRES incorporates generator-discriminator training, allowing the model to learn from more challenging examples that require structural understanding. In addition, we enrich molecular representations by transferring knowledge from scientific literature by integrating external materials embedding. Experimental results show that our model outperforms existing state-of-the-art models on popular molecular property prediction tasks.",
}
| Chemical representation learning has gained increasing interest due to the limited availability of supervised data in fields such as drug and materials design. This interest particularly extends to chemical language representation learning, which involves pre-training Transformers on SMILES sequences - textual descriptors of molecules. Despite its success in molecular property prediction, current practices often lead to overfitting and limited scalability due to early convergence. In this paper, we introduce a novel chemical language representation learning framework, called MolTRES, to address these issues. MolTRES incorporates generator-discriminator training, allowing the model to learn from more challenging examples that require structural understanding. In addition, we enrich molecular representations by transferring knowledge from scientific literature by integrating external materials embedding. Experimental results show that our model outperforms existing state-of-the-art models on popular molecular property prediction tasks. | [
"Park, Jun-Hyung",
"Kim, Yeachan",
"Lee, Mingyu",
"Park, Hyuntae",
"Lee, SangKeun"
] | MolTRES: Improving Chemical Language Representation Learning for Molecular Property Prediction | emnlp-main.788 | Poster | 2408.01426 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.789.bib | https://aclanthology.org/2024.emnlp-main.789/ | @inproceedings{aoki-etal-2024-first,
title = "First Heuristic Then Rational: Dynamic Use of Heuristics in Language Model Reasoning",
author = "Aoki, Yoichi and
Kudo, Keito and
Kuribayashi, Tatsuki and
Sone, Shusaku and
Taniguchi, Masaya and
Sakaguchi, Keisuke and
Inui, Kentaro",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.789",
pages = "14255--14271",
abstract = "Explicit multi-step reasoning, such as chain-of-thought, is widely adopted in the community to explore the better performance of language models (LMs). We report on the systematic strategy that LMs use in this process.Our controlled experiments reveal that LMs rely more heavily on heuristics, such as lexical overlap, in the earlier stages of reasoning when more steps are required to reach an answer. Conversely, their reliance on heuristics decreases as LMs progress closer to the final answer. This suggests that LMs track only a limited number of future steps and dynamically combine heuristic strategies with rational ones in solving tasks involving multi-step reasoning.",
}
| Explicit multi-step reasoning, such as chain-of-thought, is widely adopted in the community to explore the better performance of language models (LMs). We report on the systematic strategy that LMs use in this process.Our controlled experiments reveal that LMs rely more heavily on heuristics, such as lexical overlap, in the earlier stages of reasoning when more steps are required to reach an answer. Conversely, their reliance on heuristics decreases as LMs progress closer to the final answer. This suggests that LMs track only a limited number of future steps and dynamically combine heuristic strategies with rational ones in solving tasks involving multi-step reasoning. | [
"Aoki, Yoichi",
"Kudo, Keito",
"Kuribayashi, Tatsuki",
"Sone, Shusaku",
"Taniguchi, Masaya",
"Sakaguchi, Keisuke",
"Inui, Kentaro"
] | First Heuristic Then Rational: Dynamic Use of Heuristics in Language Model Reasoning | emnlp-main.789 | Poster | 2406.16078 | [
"https://github.com/ao1neko/Heuristic-and-Rational-Reasoning"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.790.bib | https://aclanthology.org/2024.emnlp-main.790/ | @inproceedings{sun-etal-2024-tools,
title = "Tools Fail: Detecting Silent Errors in Faulty Tools",
author = "Sun, Jimin and
Min, So Yeon and
Chang, Yingshan and
Bisk, Yonatan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.790",
pages = "14272--14289",
abstract = "Tools have become a mainstay of LLMs, allowing them to retrieve knowledge not in their weights, to perform tasks on the web, and even to control robots. However, most ontologies and surveys of tool-use have assumed the core challenge for LLMs is choosing the tool. Instead, we introduce a framework for tools more broadly which guides us to explore a model{'}s ability to detect {``}silent{''} tool errors, and reflect on how to plan. This more directly aligns with the increasingly popular use of models as tools. We provide an initial approach to failure recovery with promising results both on a controlled calculator setting and embodied agent planning.",
}
| Tools have become a mainstay of LLMs, allowing them to retrieve knowledge not in their weights, to perform tasks on the web, and even to control robots. However, most ontologies and surveys of tool-use have assumed the core challenge for LLMs is choosing the tool. Instead, we introduce a framework for tools more broadly which guides us to explore a model{'}s ability to detect {``}silent{''} tool errors, and reflect on how to plan. This more directly aligns with the increasingly popular use of models as tools. We provide an initial approach to failure recovery with promising results both on a controlled calculator setting and embodied agent planning. | [
"Sun, Jimin",
"Min, So Yeon",
"Chang, Yingshan",
"Bisk, Yonatan"
] | Tools Fail: Detecting Silent Errors in Faulty Tools | emnlp-main.790 | Poster | 2406.19228 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.791.bib | https://aclanthology.org/2024.emnlp-main.791/ | @inproceedings{zhang-li-2024-pcc,
title = "Pcc-tuning: Breaking the Contrastive Learning Ceiling in Semantic Textual Similarity",
author = "Zhang, Bowen and
Li, Chunping",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.791",
pages = "14290--14302",
abstract = "Semantic Textual Similarity (STS) constitutes a critical research direction in computational linguistics and serves as a key indicator of the encoding capabilities of embedding models. Driven by advances in pre-trained language models and contrastive learning, leading sentence representation methods have reached an average Spearman{'}s correlation score of approximately 86 across seven STS benchmarks in SentEval. However, further progress has become increasingly marginal, with no existing method attaining an average score higher than 86.5 on these tasks. This paper conducts an in-depth analysis of this phenomenon and concludes that the upper limit for Spearman{'}s correlation scores under contrastive learning is 87.5. To transcend this ceiling, we propose an innovative approach termed Pcc-tuning, which employs Pearson{'}s correlation coefficient as a loss function to refine model performance beyond contrastive learning. Experimental results demonstrate that Pcc-tuning can markedly surpass previous state-of-the-art strategies with only a minimal amount of fine-grained annotated samples.",
}
| Semantic Textual Similarity (STS) constitutes a critical research direction in computational linguistics and serves as a key indicator of the encoding capabilities of embedding models. Driven by advances in pre-trained language models and contrastive learning, leading sentence representation methods have reached an average Spearman{'}s correlation score of approximately 86 across seven STS benchmarks in SentEval. However, further progress has become increasingly marginal, with no existing method attaining an average score higher than 86.5 on these tasks. This paper conducts an in-depth analysis of this phenomenon and concludes that the upper limit for Spearman{'}s correlation scores under contrastive learning is 87.5. To transcend this ceiling, we propose an innovative approach termed Pcc-tuning, which employs Pearson{'}s correlation coefficient as a loss function to refine model performance beyond contrastive learning. Experimental results demonstrate that Pcc-tuning can markedly surpass previous state-of-the-art strategies with only a minimal amount of fine-grained annotated samples. | [
"Zhang, Bowen",
"Li, Chunping"
] | Pcc-tuning: Breaking the Contrastive Learning Ceiling in Semantic Textual Similarity | emnlp-main.791 | Poster | 2406.09790 | [
"https://github.com/ZBWpro/Pcc-tuning"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.792.bib | https://aclanthology.org/2024.emnlp-main.792/ | @inproceedings{kang-etal-2024-cross,
title = "Cross-lingual Back-Parsing: Utterance Synthesis from Meaning Representation for Zero-Resource Semantic Parsing",
author = "Kang, Deokhyung and
Hwang, Seonjeong and
Kim, Yunsu and
Lee, Gary",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.792",
pages = "14303--14317",
abstract = "Recent efforts have aimed to utilize multilingual pretrained language models (mPLMs) to extend semantic parsing (SP) across multiple languages without requiring extensive annotations. However, achieving zero-shot cross-lingual transfer for SP remains challenging, leading to a performance gap between source and target languages. In this study, we propose Cross-Lingual Back-Parsing (CBP), a novel data augmentation methodology designed to enhance cross-lingual transfer for SP. Leveraging the representation geometry of the mPLMs, CBP synthesizes target language utterances from source meaning representations. Our methodology effectively performs cross-lingual data augmentation in challenging zero-resource settings, by utilizing only labeled data in the source language and monolingual corpora. Extensive experiments on two cross-language SP benchmarks (Mschema2QA and Xspider) demonstrate that CBP brings substantial gains in the target language. Further analysis of the synthesized utterances shows that our method successfully generates target language utterances with high slot value alignment rates while preserving semantic integrity. Our codes and data are publicly available at https://github.com/deokhk/CBP.",
}
| Recent efforts have aimed to utilize multilingual pretrained language models (mPLMs) to extend semantic parsing (SP) across multiple languages without requiring extensive annotations. However, achieving zero-shot cross-lingual transfer for SP remains challenging, leading to a performance gap between source and target languages. In this study, we propose Cross-Lingual Back-Parsing (CBP), a novel data augmentation methodology designed to enhance cross-lingual transfer for SP. Leveraging the representation geometry of the mPLMs, CBP synthesizes target language utterances from source meaning representations. Our methodology effectively performs cross-lingual data augmentation in challenging zero-resource settings, by utilizing only labeled data in the source language and monolingual corpora. Extensive experiments on two cross-language SP benchmarks (Mschema2QA and Xspider) demonstrate that CBP brings substantial gains in the target language. Further analysis of the synthesized utterances shows that our method successfully generates target language utterances with high slot value alignment rates while preserving semantic integrity. Our codes and data are publicly available at https://github.com/deokhk/CBP. | [
"Kang, Deokhyung",
"Hwang, Seonjeong",
"Kim, Yunsu",
"Lee, Gary"
] | Cross-lingual Back-Parsing: Utterance Synthesis from Meaning Representation for Zero-Resource Semantic Parsing | emnlp-main.792 | Poster | 2410.00513 | [
"https://github.com/deokhk/cbp"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.793.bib | https://aclanthology.org/2024.emnlp-main.793/ | @inproceedings{pantazopoulos-etal-2024-shaking,
title = "Shaking Up {VLM}s: Comparing Transformers and Structured State Space Models for Vision {\&} Language Modeling",
author = "Pantazopoulos, Georgios and
Nikandrou, Malvina and
Suglia, Alessandro and
Lemon, Oliver and
Eshghi, Arash",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.793",
pages = "14318--14337",
abstract = "This study explores replacing Transformers in Visual Language Models (VLMs) with Mamba, a recent structured state space model (SSM) that demonstrates promising performance in sequence modeling. We test models up to 3B parameters under controlled conditions, showing that Mamba-based VLMs outperforms Transformers-based VLMs in captioning, question answering, and reading comprehension. However, we find that Transformers achieve greater performance in visual grounding and the performance gap widens with scale. We explore two hypotheses to explain this phenomenon: 1) the effect of task-agnostic visual encoding on the updates of the hidden states, and 2) the difficulty in performing visual grounding from the perspective of in-context multimodal retrieval. Our results indicate that a task-aware encoding yields minimal performance gains on grounding, however, Transformers significantly outperform Mamba at in-context multimodal retrieval. Overall, Mamba shows promising performance on tasks where the correct output relies on a summary of the image but struggles when retrieval of explicit information from the context is required.",
}
| This study explores replacing Transformers in Visual Language Models (VLMs) with Mamba, a recent structured state space model (SSM) that demonstrates promising performance in sequence modeling. We test models up to 3B parameters under controlled conditions, showing that Mamba-based VLMs outperforms Transformers-based VLMs in captioning, question answering, and reading comprehension. However, we find that Transformers achieve greater performance in visual grounding and the performance gap widens with scale. We explore two hypotheses to explain this phenomenon: 1) the effect of task-agnostic visual encoding on the updates of the hidden states, and 2) the difficulty in performing visual grounding from the perspective of in-context multimodal retrieval. Our results indicate that a task-aware encoding yields minimal performance gains on grounding, however, Transformers significantly outperform Mamba at in-context multimodal retrieval. Overall, Mamba shows promising performance on tasks where the correct output relies on a summary of the image but struggles when retrieval of explicit information from the context is required. | [
"Pantazopoulos, Georgios",
"Nik",
"rou, Malvina",
"Suglia, Aless",
"ro",
"Lemon, Oliver",
"Eshghi, Arash"
] | Shaking Up VLMs: Comparing Transformers and Structured State Space Models for Vision & Language Modeling | emnlp-main.793 | Poster | 2409.05395 | [
"https://github.com/gpantaz/vl_mamba"
] | https://huggingface.co/papers/2409.05395 | 1 | 0 | 0 | 5 | [] | [] | [] | [] | [] | [] | 1 |
https://aclanthology.org/2024.emnlp-main.794.bib | https://aclanthology.org/2024.emnlp-main.794/ | @inproceedings{pan-etal-2024-llms,
title = "Are {LLM}s Good Zero-Shot Fallacy Classifiers?",
author = "Pan, Fengjun and
Wu, Xiaobao and
Li, Zongrui and
Luu, Anh Tuan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.794",
pages = "14338--14364",
abstract = "Fallacies are defective arguments with faulty reasoning. Detecting and classifying them is a crucial NLP task to prevent misinformation, manipulative claims, and biased decisions. However, existing fallacy classifiers are limited by the requirement for sufficient labeled data for training, which hinders their out-of-distribution (OOD) generalization abilities. In this paper, we focus on leveraging Large Language Models (LLMs) for zero-shot fallacy classification. To elicit fallacy-related knowledge and reasoning abilities of LLMs, we propose diverse single-round and multi-round prompting schemes, applying different taskspecific instructions such as extraction, summarization, and Chain-of-Thought reasoning. With comprehensive experiments on benchmark datasets, we suggest that LLMs could be potential zero-shot fallacy classifiers. In general, LLMs under single-round prompting schemes have achieved acceptable zeroshot performances compared to the best fullshot baselines and can outperform them in all OOD inference scenarios and some opendomain tasks. Our novel multi-round prompting schemes can effectively bring about more improvements, especially for small LLMs. Our analysis further underlines the future research on zero-shot fallacy classification. Codes and data are available at: https://github.com/panFJCharlotte98/Fallacy{\_}Detection.",
}
| Fallacies are defective arguments with faulty reasoning. Detecting and classifying them is a crucial NLP task to prevent misinformation, manipulative claims, and biased decisions. However, existing fallacy classifiers are limited by the requirement for sufficient labeled data for training, which hinders their out-of-distribution (OOD) generalization abilities. In this paper, we focus on leveraging Large Language Models (LLMs) for zero-shot fallacy classification. To elicit fallacy-related knowledge and reasoning abilities of LLMs, we propose diverse single-round and multi-round prompting schemes, applying different taskspecific instructions such as extraction, summarization, and Chain-of-Thought reasoning. With comprehensive experiments on benchmark datasets, we suggest that LLMs could be potential zero-shot fallacy classifiers. In general, LLMs under single-round prompting schemes have achieved acceptable zeroshot performances compared to the best fullshot baselines and can outperform them in all OOD inference scenarios and some opendomain tasks. Our novel multi-round prompting schemes can effectively bring about more improvements, especially for small LLMs. Our analysis further underlines the future research on zero-shot fallacy classification. Codes and data are available at: https://github.com/panFJCharlotte98/Fallacy{\_}Detection. | [
"Pan, Fengjun",
"Wu, Xiaobao",
"Li, Zongrui",
"Luu, Anh Tuan"
] | Are LLMs Good Zero-Shot Fallacy Classifiers? | emnlp-main.794 | Poster | 2410.15050 | [
"https://github.com/panfjcharlotte98/fallacy_detection"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.795.bib | https://aclanthology.org/2024.emnlp-main.795/ | @inproceedings{zhou-etal-2024-mystery,
title = "The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis",
author = "Zhou, Yuxiang and
Li, Jiazheng and
Xiang, Yanzheng and
Yan, Hanqi and
Gui, Lin and
He, Yulan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.795",
pages = "14365--14378",
abstract = "Understanding in-context learning (ICL) capability that enables large language models (LLMs) to excel in proficiency through demonstration examples is of utmost importance. This importance stems not only from the better utilization of this capability across various tasks, but also from the proactive identification and mitigation of potential risks, including concerns regarding truthfulness, bias, and toxicity, that may arise alongside the capability. In this paper, we present a thorough survey on the interpretation and analysis of in-context learning. First, we provide a concise introduction to the background and definition of in-context learning. Then, we give an overview of advancements from two perspectives: 1) a theoretical perspective, emphasizing studies on mechanistic interpretability and delving into the mathematical foundations behind ICL; and 2) an empirical perspective, concerning studies that empirically analyze factors associated with ICL. We conclude by discussing open questions and the challenges encountered, and suggesting potential avenues for future research. We believe that our work establishes the basis for further exploration into the interpretation of in-context learning. To aid this effort, we have created a repository containing resources that will be continually updated.",
}
| Understanding in-context learning (ICL) capability that enables large language models (LLMs) to excel in proficiency through demonstration examples is of utmost importance. This importance stems not only from the better utilization of this capability across various tasks, but also from the proactive identification and mitigation of potential risks, including concerns regarding truthfulness, bias, and toxicity, that may arise alongside the capability. In this paper, we present a thorough survey on the interpretation and analysis of in-context learning. First, we provide a concise introduction to the background and definition of in-context learning. Then, we give an overview of advancements from two perspectives: 1) a theoretical perspective, emphasizing studies on mechanistic interpretability and delving into the mathematical foundations behind ICL; and 2) an empirical perspective, concerning studies that empirically analyze factors associated with ICL. We conclude by discussing open questions and the challenges encountered, and suggesting potential avenues for future research. We believe that our work establishes the basis for further exploration into the interpretation of in-context learning. To aid this effort, we have created a repository containing resources that will be continually updated. | [
"Zhou, Yuxiang",
"Li, Jiazheng",
"Xiang, Yanzheng",
"Yan, Hanqi",
"Gui, Lin",
"He, Yulan"
] | The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis | emnlp-main.795 | Oral | 2311.00237 | [
"https://github.com/zyxnlp/icl-interpretation-analysis-resources"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.796.bib | https://aclanthology.org/2024.emnlp-main.796/ | @inproceedings{schlechtweg-etal-2024-dwugs,
title = "More {DWUG}s: Extending and Evaluating Word Usage Graph Datasets in Multiple Languages",
author = "Schlechtweg, Dominik and
Cassotti, Pierluigi and
Noble, Bill and
Alfter, David and
Schulte Im Walde, Sabine and
Tahmasebi, Nina",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.796",
pages = "14379--14393",
abstract = "Word Usage Graphs (WUGs) represent human semantic proximity judgments for pairs of word uses in a weighted graph, which can be clustered to infer word sense clusters from simple pairwise word use judgments, avoiding the need for word sense definitions. SemEval-2020 Task 1 provided the first and to date largest manually annotated, diachronic WUG dataset. In this paper, we check the robustness and correctness of the annotations by continuing the SemEval annotation algorithm for two more rounds and comparing against an established annotation paradigm. Further, we test the reproducibility by resampling a new, smaller set of word uses from the SemEval source corpora and annotating them. Our work contributes to a better understanding of the problems and opportunities of the WUG annotation paradigm and points to future improvements.",
}
| Word Usage Graphs (WUGs) represent human semantic proximity judgments for pairs of word uses in a weighted graph, which can be clustered to infer word sense clusters from simple pairwise word use judgments, avoiding the need for word sense definitions. SemEval-2020 Task 1 provided the first and to date largest manually annotated, diachronic WUG dataset. In this paper, we check the robustness and correctness of the annotations by continuing the SemEval annotation algorithm for two more rounds and comparing against an established annotation paradigm. Further, we test the reproducibility by resampling a new, smaller set of word uses from the SemEval source corpora and annotating them. Our work contributes to a better understanding of the problems and opportunities of the WUG annotation paradigm and points to future improvements. | [
"Schlechtweg, Dominik",
"Cassotti, Pierluigi",
"Noble, Bill",
"Alfter, David",
"Schulte Im Walde, Sabine",
"Tahmasebi, Nina"
] | More DWUGs: Extending and Evaluating Word Usage Graph Datasets in Multiple Languages | emnlp-main.796 | Poster | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
||
https://aclanthology.org/2024.emnlp-main.797.bib | https://aclanthology.org/2024.emnlp-main.797/ | @inproceedings{li-etal-2024-vision,
title = "Vision-Language Model Fine-Tuning via Simple Parameter-Efficient Modification",
author = "Li, Ming and
Zhong, Jike and
Li, Chenxin and
Li, Liuzhuozheng and
Lin, Nie and
Sugiyama, Masashi",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.797",
pages = "14394--14410",
abstract = "Recent advances in fine-tuning Vision-Language Models (VLMs) have witnessed the success of prompt tuning and adapter tuning, while the classic model fine-tuning on inherent parameters seems to be overlooked. It is believed that fine-tuning the parameters of VLMs with few-shot samples corrupts the pre-trained knowledge since fine-tuning the CLIP model even degrades performance. In this paper, we revisit this viewpoint, and propose a new perspective: fine-tuning the specific parameters instead of all will uncover the power of classic model fine-tuning on VLMs. Through our meticulous study, we propose ClipFit, a simple yet effective method to fine-tune CLIP without introducing any overhead of extra parameters. We demonstrate that by only fine-tuning the specific bias terms and normalization layers, ClipFit can improve the performance of zero-shot CLIP by 7.27{\%} average harmonic mean accuracy. Lastly, to understand how fine-tuning in CLIPFit affects the pre-trained models, we conducted extensive experimental analyses w.r.t. changes in internal parameters and representations. We found that low-level text bias layers and the first layer normalization layer change much more than other layers. The code will be released.",
}
| Recent advances in fine-tuning Vision-Language Models (VLMs) have witnessed the success of prompt tuning and adapter tuning, while the classic model fine-tuning on inherent parameters seems to be overlooked. It is believed that fine-tuning the parameters of VLMs with few-shot samples corrupts the pre-trained knowledge since fine-tuning the CLIP model even degrades performance. In this paper, we revisit this viewpoint, and propose a new perspective: fine-tuning the specific parameters instead of all will uncover the power of classic model fine-tuning on VLMs. Through our meticulous study, we propose ClipFit, a simple yet effective method to fine-tune CLIP without introducing any overhead of extra parameters. We demonstrate that by only fine-tuning the specific bias terms and normalization layers, ClipFit can improve the performance of zero-shot CLIP by 7.27{\%} average harmonic mean accuracy. Lastly, to understand how fine-tuning in CLIPFit affects the pre-trained models, we conducted extensive experimental analyses w.r.t. changes in internal parameters and representations. We found that low-level text bias layers and the first layer normalization layer change much more than other layers. The code will be released. | [
"Li, Ming",
"Zhong, Jike",
"Li, Chenxin",
"Li, Liuzhuozheng",
"Lin, Nie",
"Sugiyama, Masashi"
] | Vision-Language Model Fine-Tuning via Simple Parameter-Efficient Modification | emnlp-main.797 | Poster | 2409.16718 | [
"https://github.com/minglllli/clipfit"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.798.bib | https://aclanthology.org/2024.emnlp-main.798/ | @inproceedings{phukan-etal-2024-ecis,
title = "{ECIS}-{VQG}: Generation of Entity-centric Information-seeking Questions from Videos",
author = "Phukan, Arpan and
Gupta, Manish and
Ekbal, Asif",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.798",
pages = "14411--14436",
abstract = "Previous studies on question generation from videos have mostly focused on generating questions about common objects and attributes and hence are not entity-centric. In this work, we focus on the generation of entity-centric information-seeking questions from videos. Such a system could be useful for video-based learning, recommending {``}People Also Ask{''} questions, video-based chatbots, and fact-checking. Our work addresses three key challenges: identifying question-worthy information, linking it to entities, and effectively utilizing multimodal signals. Further, to the best of our knowledge, there does not exist a large-scale dataset for this task. Most video question generation datasets are on TV shows, movies, or human activities or lack entity-centric information-seeking questions. Hence, we contribute a diverse dataset of YouTube videos, VideoQuestions, consisting of 411 videos with 2265 manually annotated questions. We further propose a model architecture combining Transformers, rich context signals (titles, transcripts, captions, embeddings), and a combination of cross-entropy and contrastive loss function to encourage entity-centric question generation. Our best method yields BLEU, ROUGE, CIDEr, and METEOR scores of 71.3, 78.6, 7.31, and 81.9, respectively, demonstrating practical usability. We make the code and dataset publicly available.",
}
| Previous studies on question generation from videos have mostly focused on generating questions about common objects and attributes and hence are not entity-centric. In this work, we focus on the generation of entity-centric information-seeking questions from videos. Such a system could be useful for video-based learning, recommending {``}People Also Ask{''} questions, video-based chatbots, and fact-checking. Our work addresses three key challenges: identifying question-worthy information, linking it to entities, and effectively utilizing multimodal signals. Further, to the best of our knowledge, there does not exist a large-scale dataset for this task. Most video question generation datasets are on TV shows, movies, or human activities or lack entity-centric information-seeking questions. Hence, we contribute a diverse dataset of YouTube videos, VideoQuestions, consisting of 411 videos with 2265 manually annotated questions. We further propose a model architecture combining Transformers, rich context signals (titles, transcripts, captions, embeddings), and a combination of cross-entropy and contrastive loss function to encourage entity-centric question generation. Our best method yields BLEU, ROUGE, CIDEr, and METEOR scores of 71.3, 78.6, 7.31, and 81.9, respectively, demonstrating practical usability. We make the code and dataset publicly available. | [
"Phukan, Arpan",
"Gupta, Manish",
"Ekbal, Asif"
] | ECIS-VQG: Generation of Entity-centric Information-seeking Questions from Videos | emnlp-main.798 | Oral | 2410.09776 | [
"https://github.com/thephukan/ecis-vqg"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.799.bib | https://aclanthology.org/2024.emnlp-main.799/ | @inproceedings{alhazmi-etal-2024-distractor,
title = "Distractor Generation in Multiple-Choice Tasks: A Survey of Methods, Datasets, and Evaluation",
author = "Alhazmi, Elaf and
Sheng, Quan Z. and
Zhang, Wei Emma and
Zaib, Munazza and
Alhazmi, Ahoud",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.799",
pages = "14437--14458",
abstract = "The distractor generation task focuses on generating incorrect but plausible options for objective questions such as fill-in-the-blank and multiple-choice questions. This task is widely utilized in educational settings across various domains and subjects. The effectiveness of these questions in assessments relies on the quality of the distractors, as they challenge examinees to select the correct answer from a set of misleading options. The evolution of artificial intelligence (AI) has transitioned the task from traditional methods to the use of neural networks and pre-trained language models. This shift has established new benchmarks and expanded the use of advanced deep learning methods in generating distractors. This survey explores distractor generation tasks, datasets, methods, and current evaluation metrics for English objective questions, covering both text-based and multi-modal domains. It also evaluates existing AI models and benchmarks and discusses potential future research directions.",
}
| The distractor generation task focuses on generating incorrect but plausible options for objective questions such as fill-in-the-blank and multiple-choice questions. This task is widely utilized in educational settings across various domains and subjects. The effectiveness of these questions in assessments relies on the quality of the distractors, as they challenge examinees to select the correct answer from a set of misleading options. The evolution of artificial intelligence (AI) has transitioned the task from traditional methods to the use of neural networks and pre-trained language models. This shift has established new benchmarks and expanded the use of advanced deep learning methods in generating distractors. This survey explores distractor generation tasks, datasets, methods, and current evaluation metrics for English objective questions, covering both text-based and multi-modal domains. It also evaluates existing AI models and benchmarks and discusses potential future research directions. | [
"Alhazmi, Elaf",
"Sheng, Quan Z.",
"Zhang, Wei Emma",
"Zaib, Munazza",
"Alhazmi, Ahoud"
] | Distractor Generation in Multiple-Choice Tasks: A Survey of Methods, Datasets, and Evaluation | emnlp-main.799 | Oral | 2402.01512 | [
""
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 0 |
|
https://aclanthology.org/2024.emnlp-main.800.bib | https://aclanthology.org/2024.emnlp-main.800/ | @inproceedings{merrill-etal-2024-evaluating,
title = "Evaluating $n$-Gram Novelty of Language Models Using Rusty-{DAWG}",
author = "Merrill, William and
Smith, Noah A. and
Elazar, Yanai",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.800",
pages = "14459--14473",
abstract = "How novel are texts generated by language models (LMs) relative to their training corpora? In this work, we investigate the extent to which modern LMs generate $n$-grams from their training data, evaluating both (i) the probability LMs assign to complete training $n$-grams and (ii) $n$-novelty, the proportion of $n$-grams generated by an LM that did not appear in the training data (for arbitrarily large $n$). To enable arbitrary-length $n$-gram search over a corpus in constant time w.r.t. corpus size, we develop Rusty-DAWG, a novel search tool inspired by indexing of genomic data. We compare the novelty of LM-generated text to human-written text and explore factors that affect generation novelty, focusing on the Pythia models. We find that, for $n > 4$, LM-generated text is less novel than human-written text, though it is more novel for smaller $n$. Larger LMs and more constrained decoding strategies both decrease novelty. Finally, we show that LMs complete $n$-grams with lower loss if they are more frequent in the training data. Overall, our results reveal factors influencing the novelty of LM-generated text, and we release Rusty-DAWG to facilitate further pretraining data research.",
}
| How novel are texts generated by language models (LMs) relative to their training corpora? In this work, we investigate the extent to which modern LMs generate $n$-grams from their training data, evaluating both (i) the probability LMs assign to complete training $n$-grams and (ii) $n$-novelty, the proportion of $n$-grams generated by an LM that did not appear in the training data (for arbitrarily large $n$). To enable arbitrary-length $n$-gram search over a corpus in constant time w.r.t. corpus size, we develop Rusty-DAWG, a novel search tool inspired by indexing of genomic data. We compare the novelty of LM-generated text to human-written text and explore factors that affect generation novelty, focusing on the Pythia models. We find that, for $n > 4$, LM-generated text is less novel than human-written text, though it is more novel for smaller $n$. Larger LMs and more constrained decoding strategies both decrease novelty. Finally, we show that LMs complete $n$-grams with lower loss if they are more frequent in the training data. Overall, our results reveal factors influencing the novelty of LM-generated text, and we release Rusty-DAWG to facilitate further pretraining data research. | [
"Merrill, William",
"Smith, Noah A.",
"Elazar, Yanai"
] | Evaluating n-Gram Novelty of Language Models Using Rusty-DAWG | emnlp-main.800 | Oral | [
"https://github.com/viking-sudo-rm/rusty-dawg"
] | -1 | -1 | -1 | -1 | [] | [] | [] | [] | [] | [] | 1 |
Subsets and Splits