publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2024-04-09T10:58:21Z | http://arxiv.org/pdf/2404.06209v1 | http://arxiv.org/abs/2404.06209v1 | Sebastian Bordt, Harsha Nori, Vanessa Rodrigues, Besmira Nushi, Rich Caruana | 2024-04-09T10:58:21Z | cs.LG, cs.AI, cs.CL | While many have shown how Large Language Models (LLMs) can be applied to a
diverse set of tasks, the critical issues of data contamination and
memorization are often glossed over. In this work, we address this concern for
tabular data. Specifically, we introduce a variety of different techniques to
assess whether a language model has seen a tabular dataset during training.
This investigation reveals that LLMs have memorized many popular tabular
datasets verbatim. We then compare the few-shot learning performance of LLMs on
datasets that were seen during training to the performance on datasets released
after training. We find that LLMs perform better on datasets seen during
training, indicating that memorization leads to overfitting. At the same time,
LLMs show non-trivial performance on novel datasets and are surprisingly robust
to data transformations. We then investigate the in-context statistical
learning abilities of LLMs. Without fine-tuning, we find them to be limited.
This suggests that much of the few-shot performance on novel datasets is due to
the LLM's world knowledge. Overall, our results highlight the importance of
testing whether an LLM has seen an evaluation dataset during pre-training. We
make the exposure tests we developed available as the tabmemcheck Python
package at https://github.com/interpretml/LLM-Tabular-Memorization-Checker | cs.LG | Elephants Never Forget: Memorization and Learning of Tabular Data in Large Language Models |
2024-04-09T10:12:34Z | http://arxiv.org/pdf/2404.06186v1 | http://arxiv.org/abs/2404.06186v1 | Andrea Zugarini, Kamyar Zeinalipour, Surya Sai Kadali, Marco Maggini, Marco Gori, Leonardo Rigutini | 2024-04-09T10:12:34Z | cs.CL, cs.AI | Crossword puzzles are popular linguistic games often used as tools to engage
students in learning. Educational crosswords are characterized by less cryptic
and more factual clues that distinguish them from traditional crossword
puzzles. Despite there exist several publicly available clue-answer pair
databases for traditional crosswords, educational clue-answer pairs datasets
are missing. In this article, we propose a methodology to build educational
clue generation datasets that can be used to instruct Large Language Models
(LLMs). By gathering from Wikipedia pages informative content associated with
relevant keywords, we use Large Language Models to automatically generate
pedagogical clues related to the given input keyword and its context. With such
an approach, we created clue-instruct, a dataset containing 44,075 unique
examples with text-keyword pairs associated with three distinct crossword
clues. We used clue-instruct to instruct different LLMs to generate educational
clues from a given input content and keyword. Both human and automatic
evaluations confirmed the quality of the generated clues, thus validating the
effectiveness of our approach. | cs.CL | Clue-Instruct: Text-Based Clue Generation for Educational Crossword Puzzles |
2024-04-09T09:34:25Z | http://arxiv.org/pdf/2404.06162v2 | http://arxiv.org/abs/2404.06162v2 | Tianyu Cao, Natraj Raman, Danial Dervovic, Chenhao Tan | 2024-05-08T04:36:07Z | cs.CL, cs.AI, cs.LG | As large language models (LLMs) expand the power of natural language
processing to handle long inputs, rigorous and systematic analyses are
necessary to understand their abilities and behavior. A salient application is
summarization, due to its ubiquity and controversy (e.g., researchers have
declared the death of summarization). In this paper, we use financial report
summarization as a case study because financial reports not only are long but
also use numbers and tables extensively. We propose a computational framework
for characterizing multimodal long-form summarization and investigate the
behavior of Claude 2.0/2.1, GPT-4/3.5, and Command. We find that GPT-3.5 and
Command fail to perform this summarization task meaningfully. For Claude 2 and
GPT-4, we analyze the extractiveness of the summary and identify a position
bias in LLMs. This position bias disappears after shuffling the input for
Claude, which suggests that Claude has the ability to recognize important
information. We also conduct a comprehensive investigation on the use of
numeric data in LLM-generated summaries and offer a taxonomy of numeric
hallucination. We employ prompt engineering to improve GPT-4's use of numbers
with limited success. Overall, our analyses highlight the strong capability of
Claude 2 in handling long multimodal inputs compared to GPT-4. | cs.CL | Characterizing Multimodal Long-form Summarization: A Case Study on Financial Reports |
2024-04-09T09:21:56Z | http://arxiv.org/pdf/2404.06150v1 | http://arxiv.org/abs/2404.06150v1 | Ben Nagy | 2024-04-09T09:21:56Z | cs.CL | This article summarizes some mostly unsuccessful attempts to understand
authorial style by examining the attention of various neural networks (LSTMs
and CNNs) trained on a corpus of classical Latin verse that has been encoded to
include sonic and metrical features. Carefully configured neural networks are
shown to be extremely strong authorship classifiers, so it is hoped that they
might therefore teach `traditional' readers something about how the authors
differ in style. Sadly their reasoning is, so far, inscrutable. While the
overall goal has not yet been reached, this work reports some useful findings
in terms of effective ways to encode and embed verse, the relative strengths
and weaknesses of the neural network families, and useful (and not so useful)
techniques for designing and inspecting NN models in this domain. This article
suggests that, for poetry, CNNs are better choices than LSTMs -- they train
more quickly, have equivalent accuracy, and (potentially) offer better
interpretability. Based on a great deal of experimentation, it also suggests
that simple, trainable embeddings are more effective than domain-specific
schemes, and stresses the importance of techniques to reduce overfitting, like
dropout and batch normalization. | cs.CL | (Not) Understanding Latin Poetic Style with Deep Learning |
2024-04-09T09:04:30Z | http://arxiv.org/pdf/2404.06138v1 | http://arxiv.org/abs/2404.06138v1 | Samuel Cahyawijaya, Holy Lovenia, Fajri Koto, Rifki Afina Putri, Emmanuel Dave, Jhonson Lee, Nuur Shadieq, Wawan Cenggoro, Salsabil Maulana Akbar, Muhammad Ihza Mahendra, Dea Annisayanti Putri, Bryan Wilie, Genta Indra Winata, Alham Fikri Aji, Ayu Purwarianti, Pascale Fung | 2024-04-09T09:04:30Z | cs.CL | Large language models (LLMs) show remarkable human-like capability in various
domains and languages. However, a notable quality gap arises in low-resource
languages, e.g., Indonesian indigenous languages, rendering them ineffective
and inefficient in such linguistic contexts. To bridge this quality gap, we
introduce Cendol, a collection of Indonesian LLMs encompassing both
decoder-only and encoder-decoder architectures across a range of model sizes.
We highlight Cendol's effectiveness across a diverse array of tasks, attaining
20% improvement, and demonstrate its capability to generalize to unseen tasks
and indigenous languages of Indonesia. Furthermore, Cendol models showcase
improved human favorability despite their limitations in capturing indigenous
knowledge and cultural values in Indonesia. In addition, we discuss the
shortcomings of parameter-efficient tunings, such as LoRA, for language
adaptation. Alternatively, we propose the usage of vocabulary adaptation to
enhance efficiency. Lastly, we evaluate the safety of Cendol and showcase that
safety in pre-training in one language such as English is transferable to
low-resource languages, such as Indonesian, even without RLHF and safety
fine-tuning. | cs.CL | Cendol: Open Instruction-tuned Generative Large Language Models for Indonesian Languages |
2024-04-09T09:03:44Z | http://arxiv.org/pdf/2404.06137v1 | http://arxiv.org/abs/2404.06137v1 | Elisei Rykov, Yana Shishkina, Kseniia Petrushina, Kseniia Titova, Sergey Petrakov, Alexander Panchenko | 2024-04-09T09:03:44Z | cs.CL, cs.AI | In this paper, we present our novel systems developed for the SemEval-2024
hallucination detection task. Our investigation spans a range of strategies to
compare model predictions with reference standards, encompassing diverse
baselines, the refinement of pre-trained encoders through supervised learning,
and an ensemble approaches utilizing several high-performing models. Through
these explorations, we introduce three distinct methods that exhibit strong
performance metrics. To amplify our training data, we generate additional
training samples from unlabelled training subset. Furthermore, we provide a
detailed comparative analysis of our approaches. Notably, our premier method
achieved a commendable 9th place in the competition's model-agnostic track and
17th place in model-aware track, highlighting its effectiveness and potential. | cs.CL | SmurfCat at SemEval-2024 Task 6: Leveraging Synthetic Data for Hallucination Detection |
2024-04-09T08:44:02Z | http://arxiv.org/pdf/2404.06121v1 | http://arxiv.org/abs/2404.06121v1 | Mariana Neves | 2024-04-09T08:44:02Z | cs.CL | Focusing on particular facts, instead of the complete text, can potentially
improve searching for specific information in the scientific literature. In
particular, argumentative elements allow focusing on specific parts of a
publication, e.g., the background section or the claims from the authors. We
evaluated some tools for the extraction of argumentation elements for a
specific task in biomedicine, namely, for detecting the fields of the
application in a biomedical publication, e.g, whether it addresses the problem
of disease diagnosis or drug development. We performed experiments with the
PubMedBERT pre-trained model, which was fine-tuned on a specific corpus for the
task. We compared the use of title and abstract to restricting to only some
argumentative elements. The top F1 scores ranged from 0.22 to 0.84, depending
on the field of application. The best argumentative labels were the ones
related the conclusion and background sections of an abstract. | cs.CL | Detection of fields of applications in biomedical abstracts with the support of argumentation elements |
2024-04-09T08:42:34Z | http://arxiv.org/pdf/2404.13057v1 | http://arxiv.org/abs/2404.13057v1 | Abhiram B. Nair, Abhinand K., Anamika U., Denil Tom Jaison, Ajitha V., V. S. Anoop | 2024-04-09T08:42:34Z | cs.CL | Sentiment analysis has become increasingly important in healthcare,
especially in the biomedical and pharmaceutical fields. The data generated by
the general public on the effectiveness, side effects, and adverse drug
reactions are goldmines for different agencies and medicine producers to
understand the concerns and reactions of people. Despite the challenge of
obtaining datasets on drug-related problems, sentiment analysis on this topic
would be a significant boon to the field. This project proposes a drug review
classification system that classifies user reviews on a particular drug into
different classes, such as positive, negative, and neutral. This approach uses
a dataset that is collected from publicly available sources containing drug
reviews, such as drugs.com. The collected data is manually labeled and verified
manually to ensure that the labels are correct. Three pre-trained language
models, such as BERT, SciBERT, and BioBERT, are used to obtain embeddings,
which were later used as features to different machine learning classifiers
such as decision trees, support vector machines, random forests, and also deep
learning algorithms such as recurrent neural networks. The performance of these
classifiers is quantified using precision, recall, and f1-score, and the
results show that the proposed approaches are useful in analyzing the
sentiments of people on different drugs. | cs.CL | "Hey..! This medicine made me sick": Sentiment Analysis of User-Generated Drug Reviews using Machine Learning Techniques |
2024-04-09T08:19:10Z | http://arxiv.org/pdf/2404.06107v1 | http://arxiv.org/abs/2404.06107v1 | Zi Long, Zhenhao Tang, Xianghua Fu, Jian Chen, Shilong Hou, Jinze Lyu | 2024-04-09T08:19:10Z | cs.CL | Recent research in the field of multimodal machine translation (MMT) has
indicated that the visual modality is either dispensable or offers only
marginal advantages. However, most of these conclusions are drawn from the
analysis of experimental results based on a limited set of bilingual
sentence-image pairs, such as Multi30k. In these kinds of datasets, the content
of one bilingual parallel sentence pair must be well represented by a manually
annotated image, which is different from the real-world translation scenario.
In this work, we adhere to the universal multimodal machine translation
framework proposed by Tang et al. (2022). This approach allows us to delve into
the impact of the visual modality on translation efficacy by leveraging
real-world translation datasets. Through a comprehensive exploration via
probing tasks, we find that the visual modality proves advantageous for the
majority of authentic translation datasets. Notably, the translation
performance primarily hinges on the alignment and coherence between textual and
visual contents. Furthermore, our results suggest that visual information
serves a supplementary role in multimodal translation and can be substituted. | cs.CL | Exploring the Necessity of Visual Modality in Multimodal Machine Translation using Authentic Datasets |
2024-04-09T08:08:03Z | http://arxiv.org/pdf/2404.06101v1 | http://arxiv.org/abs/2404.06101v1 | Blnd Yaseen, Hossein Hassani | 2024-04-09T08:08:03Z | cs.CL | Kurdish libraries have many historical publications that were printed back in
the early days when printing devices were brought to Kurdistan. Having a good
Optical Character Recognition (OCR) to help process these publications and
contribute to the Kurdish languages resources which is crucial as Kurdish is
considered a low-resource language. Current OCR systems are unable to extract
text from historical documents as they have many issues, including being
damaged, very fragile, having many marks left on them, and often written in
non-standard fonts and more. This is a massive obstacle in processing these
documents as currently processing them requires manual typing which is very
time-consuming. In this study, we adopt an open-source OCR framework by Google,
Tesseract version 5.0, that has been used to extract text for various
languages. Currently, there is no public dataset, and we developed our own by
collecting historical documents from Zheen Center for Documentation and
Research, which were printed before 1950 and resulted in a dataset of 1233
images of lines with transcription of each. Then we used the Arabic model as
our base model and trained the model using the dataset. We used different
methods to evaluate our model, Tesseracts built-in evaluator lstmeval indicated
a Character Error Rate (CER) of 0.755%. Additionally, Ocreval demonstrated an
average character accuracy of 84.02%. Finally, we developed a web application
to provide an easy- to-use interface for end-users, allowing them to interact
with the model by inputting an image of a page and extracting the text. Having
an extensive dataset is crucial to develop OCR systems with reasonable
accuracy, as currently, no public datasets are available for historical Kurdish
documents; this posed a significant challenge in our work. Additionally, the
unaligned spaces between characters and words proved another challenge with our
work. | cs.CL | Making Old Kurdish Publications Processable by Augmenting Available Optical Character Recognition Engines |
2024-04-09T08:05:15Z | http://arxiv.org/pdf/2406.15370v1 | http://arxiv.org/abs/2406.15370v1 | Sabrina Campano, Tahar Nabil, Meryl Bothua | 2024-04-09T08:05:15Z | cs.CL | This article presents a review of quantum computing research works for
Natural Language Processing (NLP). Their goal is to improve the performance of
current models, and to provide a better representation of several linguistic
phenomena, such as ambiguity and long range dependencies. Several families of
approaches are presented, including symbolic diagrammatic approaches, and
hybrid neural networks. These works show that experimental studies are already
feasible, and open research perspectives on the conception of new models and
their evaluation. | cs.CL | Traitement quantique des langues : {é}tat de l'art |
2024-04-09T07:02:14Z | http://arxiv.org/pdf/2404.06063v1 | http://arxiv.org/abs/2404.06063v1 | Baoxing Jiang | 2024-04-09T07:02:14Z | cs.CL, cs.AI, cs.LG | Aspect-Based Sentiment Analysis (ABSA) is an indispensable and highly
challenging task in natural language processing. Current efforts have focused
on specific sub-tasks, making it difficult to comprehensively cover all
sub-tasks within the ABSA domain. With the development of Generative
Pre-trained Transformers (GPTs), there came inspiration for a one-stop solution
to sentiment analysis. In this study, we used GPTs for all sub-tasks of
few-shot ABSA while defining a general learning paradigm for this application.
We propose the All in One (AiO) model, a simple yet effective two-stage model
for all ABSA sub-tasks. In the first stage, a specific backbone network learns
the semantic information of the review and generates heuristically enhanced
candidates. In the second stage, AiO leverages GPT contextual learning
capabilities to generate predictions. The study conducted comprehensive
comparative and ablation experiments on five benchmark datasets, and the
results show that AiO can effectively handle all ABSA sub-tasks, even with
few-shot data. | cs.CL | All in One: An Empirical Study of GPT for Few-Shot Aspect-Based Sentiment Anlaysis |
2024-04-09T04:55:24Z | http://arxiv.org/pdf/2404.06017v1 | http://arxiv.org/abs/2404.06017v1 | Besnik Fetahu, Nachshon Cohen, Elad Haramaty, Liane Lewin-Eytan, Oleg Rokhlenko, Shervin Malmasi | 2024-04-09T04:55:24Z | cs.CL | Voice assistants have become ubiquitous in smart devices allowing users to
instantly access information via voice questions. While extensive research has
been conducted in question answering for voice search, little attention has
been paid on how to enable proactive recommendations from a voice assistant to
its users. This is a highly challenging problem that often leads to user
friction, mainly due to recommendations provided to the users at the wrong
time. We focus on the domain of e-commerce, namely in identifying Shopping
Product Questions (SPQs), where the user asking a product-related question may
have an underlying shopping need. Identifying a user's shopping need allows
voice assistants to enhance shopping experience by determining when to provide
recommendations, such as product or deal recommendations, or proactive shopping
actions recommendation. Identifying SPQs is a challenging problem and cannot be
done from question text alone, and thus requires to infer latent user behavior
patterns inferred from user's past shopping history. We propose features that
capture the user's latent shopping behavior from their purchase history, and
combine them using a novel Mixture-of-Experts (MoE) model. Our evaluation shows
that the proposed approach is able to identify SPQs with a high score of
F1=0.91. Furthermore, based on an online evaluation with real voice assistant
users, we identify SPQs in real-time and recommend shopping actions to users to
add the queried product into their shopping list. We demonstrate that we are
able to accurately identify SPQs, as indicated by the significantly higher rate
of added products to users' shopping lists when being prompted after SPQs vs
random PQs. | cs.CL | Identifying Shopping Intent in Product QA for Proactive Recommendations |
2024-04-09T04:54:08Z | http://arxiv.org/pdf/2404.08686v1 | http://arxiv.org/abs/2404.08686v1 | Chanwoo Choi | 2024-04-09T04:54:08Z | cs.CL, cs.LG | This work demonstrates two Privacy Policy (PP) summarisation models based on
two different clustering algorithms: K-means clustering and Pre-determined
Centroid (PDC) clustering. K-means is decided to be used for the first model
after an extensive evaluation of ten commonly used clustering algorithms. The
summariser model based on the PDC-clustering algorithm summarises PP documents
by segregating individual sentences by Euclidean distance from each sentence to
the pre-defined cluster centres. The cluster centres are defined according to
General Data Protection Regulation (GDPR)'s 14 essential topics that must be
included in any privacy notices. The PDC model outperformed the K-means model
for two evaluation methods, Sum of Squared Distance (SSD) and ROUGE by some
margin (27% and 24% respectively). This result contrasts the K-means model's
better performance in the general clustering of sentence vectors before running
the task-specific evaluation. This indicates the effectiveness of operating
task-specific fine-tuning measures on unsupervised machine-learning models. The
summarisation mechanisms implemented in this paper demonstrates an idea of how
to efficiently extract essential sentences that should be included in any PP
documents. The summariser models could be further developed to an application
that tests the GDPR-compliance (or any data privacy legislation) of PP
documents. | cs.CL | Extractive text summarisation of Privacy Policy documents using machine learning approaches |
2024-04-09T04:20:27Z | http://arxiv.org/pdf/2404.06004v1 | http://arxiv.org/abs/2404.06004v1 | Kento Tatsuno, Daisuke Miyashita, Taiga Ikeda, Kiyoshi Ishiyama, Kazunari Sumiyoshi, Jun Deguchi | 2024-04-09T04:20:27Z | cs.IR, cs.CL, cs.DS | In approximate nearest neighbor search (ANNS) methods based on approximate
proximity graphs, DiskANN achieves good recall-speed balance for large-scale
datasets using both of RAM and storage. Despite it claims to save memory usage
by loading compressed vectors by product quantization (PQ), its memory usage
increases in proportion to the scale of datasets. In this paper, we propose
All-in-Storage ANNS with Product Quantization (AiSAQ), which offloads the
compressed vectors to storage. Our method achieves $\sim$10 MB memory usage in
query search even with billion-scale datasets with minor performance
degradation. AiSAQ also reduces the index load time before query search, which
enables the index switch between muitiple billion-scale datasets and
significantly enhances the flexibility of retrieval-augmented generation (RAG).
This method is applicable to all graph-based ANNS algorithms and can be
combined with higher-spec ANNS methods in the future. | cs.IR | AiSAQ: All-in-Storage ANNS with Product Quantization for DRAM-free Information Retrieval |
2024-04-09T04:17:51Z | http://arxiv.org/pdf/2404.06003v1 | http://arxiv.org/abs/2404.06003v1 | Zhuohao Yu, Chang Gao, Wenjin Yao, Yidong Wang, Zhengran Zeng, Wei Ye, Jindong Wang, Yue Zhang, Shikun Zhang | 2024-04-09T04:17:51Z | cs.CL, cs.AI | The rapid development of large language model (LLM) evaluation methodologies
and datasets has led to a profound challenge: integrating state-of-the-art
evaluation techniques cost-effectively while ensuring reliability,
reproducibility, and efficiency. Currently, there is a notable absence of a
unified and adaptable framework that seamlessly integrates various evaluation
approaches. Moreover, the reliability of evaluation findings is often
questionable due to potential data contamination, with the evaluation
efficiency commonly overlooked when facing the substantial costs associated
with LLM inference. In response to these challenges, we introduce FreeEval, a
modular and scalable framework crafted to enable trustworthy and efficient
automatic evaluations of LLMs. Firstly, FreeEval's unified abstractions
simplify the integration and improve the transparency of diverse evaluation
methodologies, encompassing dynamic evaluation that demand sophisticated LLM
interactions. Secondly, the framework integrates meta-evaluation techniques
like human evaluation and data contamination detection, which, along with
dynamic evaluation modules in the platform, enhance the fairness of the
evaluation outcomes. Lastly, FreeEval is designed with a high-performance
infrastructure, including distributed computation and caching strategies,
enabling extensive evaluations across multi-node, multi-GPU clusters for
open-source and proprietary LLMs. | cs.CL | FreeEval: A Modular Framework for Trustworthy and Efficient Evaluation of Large Language Models |
2024-04-09T04:11:25Z | http://arxiv.org/pdf/2404.06001v2 | http://arxiv.org/abs/2404.06001v2 | Kennedy Edemacu, Xintao Wu | 2024-04-11T00:17:18Z | cs.CL | Pre-trained language models (PLMs) have demonstrated significant proficiency
in solving a wide range of general natural language processing (NLP) tasks.
Researchers have observed a direct correlation between the performance of these
models and their sizes. As a result, the sizes of these models have notably
expanded in recent years, persuading researchers to adopt the term large
language models (LLMs) to characterize the larger-sized PLMs. The size
expansion comes with a distinct capability called in-context learning (ICL),
which represents a special form of prompting and allows the models to be
utilized through the presentation of demonstration examples without
modifications to the model parameters. Although interesting, privacy concerns
have become a major obstacle in its widespread usage. Multiple studies have
examined the privacy risks linked to ICL and prompting in general, and have
devised techniques to alleviate these risks. Thus, there is a necessity to
organize these mitigation techniques for the benefit of the community. This
survey provides a systematic overview of the privacy protection methods
employed during ICL and prompting in general. We review, analyze, and compare
different methods under this paradigm. Furthermore, we provide a summary of the
resources accessible for the development of these frameworks. Finally, we
discuss the limitations of these frameworks and offer a detailed examination of
the promising areas that necessitate further exploration. | cs.CL | Privacy Preserving Prompt Engineering: A Survey |
2024-04-09T03:54:28Z | http://arxiv.org/pdf/2404.05993v1 | http://arxiv.org/abs/2404.05993v1 | Shaona Ghosh, Prasoon Varshney, Erick Galinkin, Christopher Parisien | 2024-04-09T03:54:28Z | cs.LG, cs.CL, cs.CY | As Large Language Models (LLMs) and generative AI become more widespread, the
content safety risks associated with their use also increase. We find a notable
deficiency in high-quality content safety datasets and benchmarks that
comprehensively cover a wide range of critical safety areas. To address this,
we define a broad content safety risk taxonomy, comprising 13 critical risk and
9 sparse risk categories. Additionally, we curate AEGISSAFETYDATASET, a new
dataset of approximately 26, 000 human-LLM interaction instances, complete with
human annotations adhering to the taxonomy. We plan to release this dataset to
the community to further research and to help benchmark LLM models for safety.
To demonstrate the effectiveness of the dataset, we instruction-tune multiple
LLM-based safety models. We show that our models (named AEGISSAFETYEXPERTS),
not only surpass or perform competitively with the state-of-the-art LLM-based
safety models and general purpose LLMs, but also exhibit robustness across
multiple jail-break attack categories. We also show how using
AEGISSAFETYDATASET during the LLM alignment phase does not negatively impact
the performance of the aligned models on MT Bench scores. Furthermore, we
propose AEGIS, a novel application of a no-regret online adaptation framework
with strong theoretical guarantees, to perform content moderation with an
ensemble of LLM content safety experts in deployment | cs.LG | AEGIS: Online Adaptive AI Content Safety Moderation with Ensemble of LLM Experts |
2024-04-09T03:47:48Z | http://arxiv.org/pdf/2404.05989v1 | http://arxiv.org/abs/2404.05989v1 | Yanan Zhang, Xiaoling Bai, Tianhua Zhou | 2024-04-09T03:47:48Z | cs.CL, cs.IR | The embedding-based retrieval (EBR) approach is widely used in mainstream
search engine retrieval systems and is crucial in recent retrieval-augmented
methods for eliminating LLM illusions. However, existing EBR models often face
the "semantic drift" problem and insufficient focus on key information, leading
to a low adoption rate of retrieval results in subsequent steps. This issue is
especially noticeable in real-time search scenarios, where the various
expressions of popular events on the Internet make real-time retrieval heavily
reliant on crucial event information. To tackle this problem, this paper
proposes a novel approach called EER, which enhances real-time retrieval
performance by improving the dual-encoder model of traditional EBR. We
incorporate contrastive learning to accompany pairwise learning for encoder
optimization. Furthermore, to strengthen the focus on critical event
information in events, we include a decoder module after the document encoder,
introduce a generative event triplet extraction scheme based on prompt-tuning,
and correlate the events with query encoder optimization through comparative
learning. This decoder module can be removed during inference. Extensive
experiments demonstrate that EER can significantly improve the real-time search
retrieval performance. We believe that this approach will provide new
perspectives in the field of information retrieval. The codes and dataset are
available at https://github.com/open-event-hub/Event-enhanced_Retrieval . | cs.CL | Event-enhanced Retrieval in Real-time Search |
2024-04-09T02:59:17Z | http://arxiv.org/pdf/2404.05971v1 | http://arxiv.org/abs/2404.05971v1 | Gonçalo Paulo, Thomas Marshall, Nora Belrose | 2024-04-09T02:59:17Z | cs.LG, cs.AI, cs.CL | Recent advances in recurrent neural network architectures, such as Mamba and
RWKV, have enabled RNNs to match or exceed the performance of equal-size
transformers in terms of language modeling perplexity and downstream
evaluations, suggesting that future systems may be built on completely new
architectures. In this paper, we examine if selected interpretability methods
originally designed for transformer language models will transfer to these
up-and-coming recurrent architectures. Specifically, we focus on steering model
outputs via contrastive activation addition, on eliciting latent predictions
via the tuned lens, and eliciting latent knowledge from models fine-tuned to
produce false outputs under certain conditions. Our results show that most of
these techniques are effective when applied to RNNs, and we show that it is
possible to improve some of them by taking advantage of RNNs' compressed state. | cs.LG | Does Transformer Interpretability Transfer to RNNs? |
2024-04-09T02:58:05Z | http://arxiv.org/pdf/2404.05970v1 | http://arxiv.org/abs/2404.05970v1 | Alireza Salemi, Surya Kallumadi, Hamed Zamani | 2024-04-09T02:58:05Z | cs.CL, cs.IR | This paper studies retrieval-augmented approaches for personalizing large
language models (LLMs), which potentially have a substantial impact on various
applications and domains. We propose the first attempt to optimize the
retrieval models that deliver a limited number of personal documents to large
language models for the purpose of personalized generation. We develop two
optimization algorithms that solicit feedback from the downstream personalized
generation tasks for retrieval optimization -- one based on reinforcement
learning whose reward function is defined using any arbitrary metric for
personalized generation and another based on knowledge distillation from the
downstream LLM to the retrieval model. This paper also introduces a pre- and
post-generation retriever selection model that decides what retriever to choose
for each LLM input. Extensive experiments on diverse tasks from the language
model personalization (LaMP) benchmark reveal statistically significant
improvements in six out of seven datasets. | cs.CL | Optimization Methods for Personalizing Large Language Models through Retrieval Augmentation |
2024-04-09T02:55:12Z | http://arxiv.org/pdf/2404.05967v1 | http://arxiv.org/abs/2404.05967v1 | Masato Fujitake | 2024-04-09T02:55:12Z | cs.CV, cs.AI, cs.CL, cs.LG | In this paper, we present a method for enhancing the accuracy of scene text
recognition tasks by judging whether the image and text match each other. While
previous studies focused on generating the recognition results from input
images, our approach also considers the model's misrecognition results to
understand its error tendencies, thus improving the text recognition pipeline.
This method boosts text recognition accuracy by providing explicit feedback on
the data that the model is likely to misrecognize by predicting correct or
incorrect between the image and text. The experimental results on publicly
available datasets demonstrate that our proposed method outperforms the
baseline and state-of-the-art methods in scene text recognition. | cs.CV | JSTR: Judgment Improves Scene Text Recognition |
2024-04-09T02:53:14Z | http://arxiv.org/pdf/2404.05966v1 | http://arxiv.org/abs/2404.05966v1 | Yizhou Chi, Kevin Yang, Dan Klein | 2024-04-09T02:53:14Z | cs.CL, cs.AI | We present THOUGHTSCULPT, a general reasoning and search method for tasks
with outputs that can be decomposed into components. THOUGHTSCULPT explores a
search tree of potential solutions using Monte Carlo Tree Search (MCTS),
building solutions one action at a time and evaluating according to any
domain-specific heuristic, which in practice is often simply an LLM evaluator.
Critically, our action space includes revision actions: THOUGHTSCULPT may
choose to revise part of its previous output rather than continuing to build
the rest of its output. Empirically, THOUGHTSCULPT outperforms state-of-the-art
reasoning methods across three challenging tasks: Story Outline Improvement (up
to +30% interestingness), Mini-Crosswords Solving (up to +16% word success
rate), and Constrained Generation (up to +10% concept coverage). | cs.CL | THOUGHTSCULPT: Reasoning with Intermediate Revision and Search |
2024-04-09T02:51:05Z | http://arxiv.org/pdf/2404.05961v1 | http://arxiv.org/abs/2404.05961v1 | Parishad BehnamGhader, Vaibhav Adlakha, Marius Mosbach, Dzmitry Bahdanau, Nicolas Chapados, Siva Reddy | 2024-04-09T02:51:05Z | cs.CL, cs.AI | Large decoder-only language models (LLMs) are the state-of-the-art models on
most of today's NLP tasks and benchmarks. Yet, the community is only slowly
adopting these models for text embedding tasks, which require rich
contextualized representations. In this work, we introduce LLM2Vec, a simple
unsupervised approach that can transform any decoder-only LLM into a strong
text encoder. LLM2Vec consists of three simple steps: 1) enabling bidirectional
attention, 2) masked next token prediction, and 3) unsupervised contrastive
learning. We demonstrate the effectiveness of LLM2Vec by applying it to 3
popular LLMs ranging from 1.3B to 7B parameters and evaluate the transformed
models on English word- and sequence-level tasks. We outperform encoder-only
models by a large margin on word-level tasks and reach a new unsupervised
state-of-the-art performance on the Massive Text Embeddings Benchmark (MTEB).
Moreover, when combining LLM2Vec with supervised contrastive learning, we
achieve state-of-the-art performance on MTEB among models that train only on
publicly available data. Our strong empirical results and extensive analysis
demonstrate that LLMs can be effectively transformed into universal text
encoders in a parameter-efficient manner without the need for expensive
adaptation or synthetic GPT-4 generated data. | cs.CL | LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders |
2024-04-09T02:29:39Z | http://arxiv.org/pdf/2404.05955v1 | http://arxiv.org/abs/2404.05955v1 | Junpeng Liu, Yifan Song, Bill Yuchen Lin, Wai Lam, Graham Neubig, Yuanzhi Li, Xiang Yue | 2024-04-09T02:29:39Z | cs.CL, cs.AI | Multimodal Large Language models (MLLMs) have shown promise in web-related
tasks, but evaluating their performance in the web domain remains a challenge
due to the lack of comprehensive benchmarks. Existing benchmarks are either
designed for general multimodal tasks, failing to capture the unique
characteristics of web pages, or focus on end-to-end web agent tasks, unable to
measure fine-grained abilities such as OCR, understanding, and grounding. In
this paper, we introduce \bench{}, a multimodal benchmark designed to assess
the capabilities of MLLMs across a variety of web tasks. \bench{} consists of
seven tasks, and comprises 1.5K human-curated instances from 139 real websites,
covering 87 sub-domains. We evaluate 14 open-source MLLMs, Gemini Pro, Claude-3
series, and GPT-4V(ision) on \bench{}, revealing significant challenges and
performance gaps. Further analysis highlights the limitations of current MLLMs,
including inadequate grounding in text-rich environments and subpar performance
with low-resolution image inputs. We believe \bench{} will serve as a valuable
resource for the research community and contribute to the creation of more
powerful and versatile MLLMs for web-related applications. | cs.CL | VisualWebBench: How Far Have Multimodal LLMs Evolved in Web Page Understanding and Grounding? |
2024-04-09T01:55:05Z | http://arxiv.org/pdf/2404.05943v1 | http://arxiv.org/abs/2404.05943v1 | Wei-Rui Chen, Ife Adebara, Muhammad Abdul-Mageed | 2024-04-09T01:55:05Z | cs.CL, cs.AI | We investigate two research questions: (1) how do machine translation (MT)
and diacritization influence the performance of each other in a multi-task
learning setting (2) the effect of keeping (vs. removing) diacritics on MT
performance. We examine these two questions in both high-resource (HR) and
low-resource (LR) settings across 55 different languages (36 African languages
and 19 European languages). For (1), results show that diacritization
significantly benefits MT in the LR scenario, doubling or even tripling
performance for some languages, but harms MT in the HR scenario. We find that
MT harms diacritization in LR but benefits significantly in HR for some
languages. For (2), MT performance is similar regardless of diacritics being
kept or removed. In addition, we propose two classes of metrics to measure the
complexity of a diacritical system, finding these metrics to correlate
positively with the performance of our diacritization models. Overall, our work
provides insights for developing MT and diacritization systems under different
data size conditions and may have implications that generalize beyond the 55
languages we investigate. | cs.CL | Interplay of Machine Translation, Diacritics, and Diacritization |
2024-04-08T23:16:22Z | http://arxiv.org/pdf/2404.05904v2 | http://arxiv.org/abs/2404.05904v2 | Giwon Hong, Aryo Pradipta Gema, Rohit Saxena, Xiaotang Du, Ping Nie, Yu Zhao, Laura Perez-Beltrachini, Max Ryabinin, Xuanli He, Clémentine Fourrier, Pasquale Minervini | 2024-04-17T07:03:17Z | cs.CL | Large Language Models (LLMs) have transformed the Natural Language Processing
(NLP) landscape with their remarkable ability to understand and generate
human-like text. However, these models are prone to ``hallucinations'' --
outputs that do not align with factual reality or the input context. This paper
introduces the Hallucinations Leaderboard, an open initiative to quantitatively
measure and compare the tendency of each model to produce hallucinations. The
leaderboard uses a comprehensive set of benchmarks focusing on different
aspects of hallucinations, such as factuality and faithfulness, across various
tasks, including question-answering, summarisation, and reading comprehension.
Our analysis provides insights into the performance of different models,
guiding researchers and practitioners in choosing the most reliable models for
their applications. | cs.CL | The Hallucinations Leaderboard -- An Open Effort to Measure Hallucinations in Large Language Models |
2024-04-08T23:10:47Z | http://arxiv.org/pdf/2404.05902v1 | http://arxiv.org/abs/2404.05902v1 | Michael Lutz, Arth Bohra, Manvel Saroyan, Artem Harutyunyan, Giovanni Campagna | 2024-04-08T23:10:47Z | cs.CL, cs.AI | In the realm of web agent research, achieving both generalization and
accuracy remains a challenging problem. Due to high variance in website
structure, existing approaches often fail. Moreover, existing fine-tuning and
in-context learning techniques fail to generalize across multiple websites. We
introduce Wilbur, an approach that uses a differentiable ranking model and a
novel instruction synthesis technique to optimally populate a black-box large
language model's prompt with task demonstrations from previous runs. To
maximize end-to-end success rates, we also propose an intelligent backtracking
mechanism that learns and recovers from its mistakes. Finally, we show that our
ranking model can be trained on data from a generative auto-curriculum which
samples representative goals from an LLM, runs the agent, and automatically
evaluates it, with no manual annotation. Wilbur achieves state-of-the-art
results on the WebVoyager benchmark, beating text-only models by 8% overall,
and up to 36% on certain websites. On the same benchmark, Wilbur is within 5%
of a strong multi-modal model despite only receiving textual inputs, and
further analysis reveals a substantial number of failures are due to
engineering challenges of operating the web. | cs.CL | WILBUR: Adaptive In-Context Learning for Robust and Accurate Web Agents |
2024-04-08T22:29:53Z | http://arxiv.org/pdf/2404.05893v2 | http://arxiv.org/abs/2404.05893v2 | Sowmya S. Sundaram, Benjamin Solomon, Avani Khatri, Anisha Laumas, Purvesh Khatri, Mark A. Musen | 2024-04-17T04:17:12Z | cs.AI, cs.CL, cs.IR | Metadata play a crucial role in ensuring the findability, accessibility,
interoperability, and reusability of datasets. This paper investigates the
potential of large language models (LLMs), specifically GPT-4, to improve
adherence to metadata standards. We conducted experiments on 200 random data
records describing human samples relating to lung cancer from the NCBI
BioSample repository, evaluating GPT-4's ability to suggest edits for adherence
to metadata standards. We computed the adherence accuracy of field name-field
value pairs through a peer review process, and we observed a marginal average
improvement in adherence to the standard data dictionary from 79% to 80%
(p<0.01). We then prompted GPT-4 with domain information in the form of the
textual descriptions of CEDAR templates and recorded a significant improvement
to 97% from 79% (p<0.01). These results indicate that, while LLMs may not be
able to correct legacy metadata to ensure satisfactory adherence to standards
when unaided, they do show promise for use in automated metadata curation when
integrated with a structured knowledge base. | cs.AI | Use of a Structured Knowledge Base Enhances Metadata Curation by Large Language Models |
2024-04-08T22:20:59Z | http://arxiv.org/pdf/2404.05892v2 | http://arxiv.org/abs/2404.05892v2 | Bo Peng, Daniel Goldstein, Quentin Anthony, Alon Albalak, Eric Alcaide, Stella Biderman, Eugene Cheah, Xingjian Du, Teddy Ferdinan, Haowen Hou, Przemysław Kazienko, Kranthi Kiran GV, Jan Kocoń, Bartłomiej Koptyra, Satyapriya Krishna, Ronald McClelland Jr., Niklas Muennighoff, Fares Obeid, Atsushi Saito, Guangyu Song, Haoqin Tu, Stanisław Woźniak, Ruichong Zhang, Bingchen Zhao, Qihang Zhao, Peng Zhou, Jian Zhu, Rui-Jie Zhu | 2024-04-10T19:34:38Z | cs.CL, cs.AI | We present Eagle (RWKV-5) and Finch (RWKV-6), sequence models improving upon
the RWKV (RWKV-4) architecture. Our architectural design advancements include
multi-headed matrix-valued states and a dynamic recurrence mechanism that
improve expressivity while maintaining the inference efficiency characteristics
of RNNs. We introduce a new multilingual corpus with 1.12 trillion tokens and a
fast tokenizer based on greedy matching for enhanced multilinguality. We
trained four Eagle models, ranging from 0.46 to 7.5 billion parameters, and two
Finch models with 1.6 and 3.1 billion parameters and find that they achieve
competitive performance across a wide variety of benchmarks. We release all our
models on HuggingFace under the Apache 2.0 license. Models at:
https://huggingface.co/RWKV Training code at: https://github.com/RWKV/RWKV-LM
Inference code at: https://github.com/RWKV/ChatRWKV Time-parallel training code
at: https://github.com/RWKV/RWKV-infctx-trainer | cs.CL | Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence |
2024-04-08T21:26:22Z | http://arxiv.org/pdf/2404.05880v2 | http://arxiv.org/abs/2404.05880v2 | Weikai Lu, Ziqian Zeng, Jianwei Wang, Zhengdong Lu, Zelin Chen, Huiping Zhuang, Cen Chen | 2024-07-03T17:52:00Z | cs.CL | Jailbreaking attacks can enable Large Language Models (LLMs) to bypass the
safeguard and generate harmful content. Existing jailbreaking defense methods
have failed to address the fundamental issue that harmful knowledge resides
within the model, leading to potential jailbreak risks for LLMs. In this paper,
we propose a novel defense method called Eraser, which mainly includes three
goals: unlearning harmful knowledge, retaining general knowledge, and
maintaining safety alignment. The intuition is that if an LLM forgets the
specific knowledge required to answer a harmful question, it will no longer
have the ability to answer harmful questions. The training of Erase does not
actually require the model's own harmful knowledge, and it can benefit from
unlearning general answers related to harmful queries, which means it does not
need assistance from the red team. The experimental results show that Eraser
can significantly reduce the jailbreaking success rate for various attacks
without compromising the general capabilities of the model. Our codes are
available at https://github.com/ZeroNLP/Eraser. | cs.CL | Eraser: Jailbreaking Defense in Large Language Models via Unlearning Harmful Knowledge |
2024-04-08T21:15:36Z | http://arxiv.org/pdf/2404.05875v1 | http://arxiv.org/abs/2404.05875v1 | Zifeng Wang, Chun-Liang Li, Vincent Perot, Long T. Le, Jin Miao, Zizhao Zhang, Chen-Yu Lee, Tomas Pfister | 2024-04-08T21:15:36Z | cs.CL, cs.AI, cs.LG | Instruction tuning has emerged as the key in aligning large language models
(LLMs) with specific task instructions, thereby mitigating the discrepancy
between the next-token prediction objective and users' actual goals. To reduce
the labor and time cost to collect or annotate data by humans, researchers
start to explore the use of LLMs to generate instruction-aligned synthetic
data. Recent works focus on generating diverse instructions and applying LLM to
increase instruction complexity, often neglecting downstream use cases. It
remains unclear how to tailor high-quality data to elicit better
instruction-following abilities in different target instruction distributions
and LLMs. To this end, we introduce CodecLM, a general framework for adaptively
generating high-quality synthetic data for LLM alignment with different
downstream instruction distributions and LLMs. Drawing on the Encode-Decode
principles, we use LLMs as codecs to guide the data generation process. We
first encode seed instructions into metadata, which are concise keywords
generated on-the-fly to capture the target instruction distribution, and then
decode metadata to create tailored instructions. We also introduce Self-Rubrics
and Contrastive Filtering during decoding to tailor data-efficient samples.
Extensive experiments on four open-domain instruction following benchmarks
validate the effectiveness of CodecLM over the current state-of-the-arts. | cs.CL | CodecLM: Aligning Language Models with Tailored Synthetic Data |
2024-04-08T21:05:42Z | http://arxiv.org/pdf/2404.05868v1 | http://arxiv.org/abs/2404.05868v1 | Ruiqi Zhang, Licong Lin, Yu Bai, Song Mei | 2024-04-08T21:05:42Z | cs.LG, cs.AI, cs.CL, stat.ML | Large Language Models (LLMs) often memorize sensitive, private, or
copyrighted data during pre-training. LLM unlearning aims to eliminate the
influence of undesirable data from the pre-trained model while preserving the
model's utilities on other tasks. Several practical methods have recently been
proposed for LLM unlearning, mostly based on gradient ascent (GA) on the loss
of undesirable data. However, on certain unlearning tasks, these methods either
fail to effectively unlearn the target data or suffer from catastrophic
collapse -- a drastic degradation of the model's utilities.
In this paper, we propose Negative Preference Optimization (NPO), a simple
alignment-inspired method that could efficiently and effectively unlearn a
target dataset. We theoretically show that the progression toward catastrophic
collapse by minimizing the NPO loss is exponentially slower than GA. Through
experiments on synthetic data and the benchmark TOFU dataset, we demonstrate
that NPO-based methods achieve a better balance between unlearning the
undesirable data and maintaining the model's utilities. We also observe that
NPO-based methods generate more sensible outputs than GA-based methods, whose
outputs are often gibberish. Remarkably, on TOFU, NPO-based methods are the
first to achieve reasonable unlearning results in forgetting 50% (or more) of
the training data, whereas existing methods already struggle with forgetting
10% of training data. | cs.LG | Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning |
2024-04-08T20:58:06Z | http://arxiv.org/pdf/2404.05866v1 | http://arxiv.org/abs/2404.05866v1 | Aida Mostafazadeh Davani, Sagar Gubbi, Sunipa Dev, Shachi Dave, Vinodkumar Prabhakaran | 2024-04-08T20:58:06Z | cs.CL | LLMs are increasingly transforming our digital ecosystem, but they often
inherit societal biases learned from their training data, for instance
stereotypes associating certain attributes with specific identity groups. While
whether and how these biases are mitigated may depend on the specific use
cases, being able to effectively detect instances of stereotype perpetuation is
a crucial first step. Current methods to assess presence of stereotypes in
generated language rely on simple template or co-occurrence based measures,
without accounting for the variety of sentential contexts they manifest in. We
argue that understanding the sentential context is crucial for detecting
instances of generalization. We distinguish two types of generalizations: (1)
language that merely mentions the presence of a generalization ("people think
the French are very rude"), and (2) language that reinforces such a
generalization ("as French they must be rude"), from non-generalizing context
("My French friends think I am rude"). For meaningful stereotype evaluations,
we need to reliably distinguish such instances of generalizations. We introduce
the new task of detecting generalization in language, and build GeniL, a
multilingual dataset of over 50K sentences from 9 languages (English, Arabic,
Bengali, Spanish, French, Hindi, Indonesian, Malay, and Portuguese) annotated
for instances of generalizations. We demonstrate that the likelihood of a
co-occurrence being an instance of generalization is usually low, and varies
across different languages, identity groups, and attributes. We build
classifiers to detect generalization in language with an overall PR-AUC of
58.7, with varying degrees of performance across languages. Our research
provides data and tools to enable a nuanced understanding of stereotype
perpetuation, a crucial step towards more inclusive and responsible language
technologies. | cs.CL | GeniL: A Multilingual Dataset on Generalizing Language |
2024-04-08T20:14:10Z | http://arxiv.org/pdf/2404.05843v2 | http://arxiv.org/abs/2404.05843v2 | Franz A. Heinsen | 2024-04-27T19:03:14Z | cs.LG, cs.CL | We propose a simple modification to the conventional attention mechanism
applied by Transformers: Instead of quantifying pairwise query-key similarity
with scaled dot-products, we quantify it with the logarithms of scaled
dot-products of exponentials. Our modification linearizes attention with
exponential kernel feature maps, whose corresponding feature function is
infinite dimensional. We show that our modification is expressible as a
composition of log-sums of exponentials, with a latent space of constant size,
enabling application with constant time and space complexity per token. We
implement our modification, verify that it works in practice, and conclude that
it is a promising alternative to conventional attention. | cs.LG | Softmax Attention with Constant Cost per Token |
2024-04-08T20:05:25Z | http://arxiv.org/pdf/2404.05839v2 | http://arxiv.org/abs/2404.05839v2 | Milan Straka, Jana Straková, Federica Gamba | 2024-05-29T16:37:39Z | cs.CL | We present LatinPipe, the winning submission to the EvaLatin 2024 Dependency
Parsing shared task. Our system consists of a fine-tuned concatenation of base
and large pre-trained LMs, with a dot-product attention head for parsing and
softmax classification heads for morphology to jointly learn both dependency
parsing and morphological analysis. It is trained by sampling from seven
publicly available Latin corpora, utilizing additional harmonization of
annotations to achieve a more unified annotation style. Before fine-tuning, we
train the system for a few initial epochs with frozen weights. We also add
additional local relative contextualization by stacking the BiLSTM layers on
top of the Transformer(s). Finally, we ensemble output probability
distributions from seven randomly instantiated networks for the final
submission. The code is available at
https://github.com/ufal/evalatin2024-latinpipe. | cs.CL | ÚFAL LatinPipe at EvaLatin 2024: Morphosyntactic Analysis of Latin |
2024-04-08T19:48:36Z | http://arxiv.org/pdf/2404.05829v1 | http://arxiv.org/abs/2404.05829v1 | Zoltan Csaki, Bo Li, Jonathan Li, Qiantong Xu, Pian Pawakapan, Leon Zhang, Yun Du, Hengyu Zhao, Changran Hu, Urmish Thakker | 2024-04-08T19:48:36Z | cs.CL, cs.AI, cs.LG | Despite the widespread availability of LLMs, there remains a substantial gap
in their capabilities and availability across diverse languages. One approach
to address these issues has been to take an existing pre-trained LLM and
continue to train it on new languages. While prior works have experimented with
language adaptation, many questions around best practices and methodology have
not been covered. In this paper, we present a comprehensive investigation into
the adaptation of LLMs to new languages. Our study covers the key components in
this process, including vocabulary extension, direct preference optimization
and the data scarcity problem for human alignment in low-resource languages. We
scale these experiments across 9 languages and 2 parameter scales (7B and 70B).
We compare our models against Llama 2, Aya-101, XGLM, BLOOM and existing
language experts, outperforming all prior published baselines. Additionally,
all evaluation code and checkpoints are made public to facilitate future
research. | cs.CL | SambaLingo: Teaching Large Language Models New Languages |
2024-04-08T18:33:59Z | http://arxiv.org/pdf/2404.08685v1 | http://arxiv.org/abs/2404.08685v1 | Bhavith Chandra Challagundla, Chakradhar Peddavenkatagari | 2024-04-08T18:33:59Z | cs.CL, cs.LG | Automatic text summarization (TS) plays a pivotal role in condensing large
volumes of information into concise, coherent summaries, facilitating efficient
information retrieval and comprehension. This paper presents a novel framework
for abstractive TS of single documents, which integrates three dominant
aspects: structural, semantic, and neural-based approaches. The proposed
framework merges machine learning and knowledge-based techniques to achieve a
unified methodology. The framework consists of three main phases:
pre-processing, machine learning, and post-processing. In the pre-processing
phase, a knowledge-based Word Sense Disambiguation (WSD) technique is employed
to generalize ambiguous words, enhancing content generalization. Semantic
content generalization is then performed to address out-of-vocabulary (OOV) or
rare words, ensuring comprehensive coverage of the input document.
Subsequently, the generalized text is transformed into a continuous vector
space using neural language processing techniques. A deep sequence-to-sequence
(seq2seq) model with an attention mechanism is employed to predict a
generalized summary based on the vector representation. In the post-processing
phase, heuristic algorithms and text similarity metrics are utilized to refine
the generated summary further. Concepts from the generalized summary are
matched with specific entities, enhancing coherence and readability.
Experimental evaluations conducted on prominent datasets, including Gigaword,
Duc 2004, and CNN/DailyMail, demonstrate the effectiveness of the proposed
framework. Results indicate significant improvements in handling rare and OOV
words, outperforming existing state-of-the-art deep learning techniques. The
proposed framework presents a comprehensive and unified approach towards
abstractive TS, combining the strengths of structure, semantics, and
neural-based methodologies. | cs.CL | Neural Sequence-to-Sequence Modeling with Attention by Leveraging Deep Learning Architectures for Enhanced Contextual Understanding in Abstractive Text Summarization |
2024-04-08T17:56:43Z | http://arxiv.org/pdf/2404.05720v1 | http://arxiv.org/abs/2404.05720v1 | Vladimir Solovyev, Danni Liu, Jan Niehues | 2024-04-08T17:56:43Z | cs.CL, cs.AI | Finetuning pretrained models on downstream generation tasks often leads to
catastrophic forgetting in zero-shot conditions. In this work, we focus on
summarization and tackle the problem through the lens of language-independent
representations. After training on monolingual summarization, we perform
zero-shot transfer to new languages or language pairs. We first show naively
finetuned models are highly language-specific in both output behavior and
internal representations, resulting in poor zero-shot performance. Next, we
propose query-key (QK) finetuning to decouple task-specific knowledge from the
pretrained language generation abilities. Then, after showing downsides of the
standard adversarial language classifier, we propose a balanced variant that
more directly enforces language-agnostic representations. Moreover, our
qualitative analyses show removing source language identity correlates to
zero-shot summarization performance. Our code is openly available. | cs.CL | Language-Independent Representations Improve Zero-Shot Summarization |
2024-04-08T17:55:44Z | http://arxiv.org/pdf/2404.05719v1 | http://arxiv.org/abs/2404.05719v1 | Keen You, Haotian Zhang, Eldon Schoop, Floris Weers, Amanda Swearngin, Jeffrey Nichols, Yinfei Yang, Zhe Gan | 2024-04-08T17:55:44Z | cs.CV, cs.CL, cs.HC | Recent advancements in multimodal large language models (MLLMs) have been
noteworthy, yet, these general-domain MLLMs often fall short in their ability
to comprehend and interact effectively with user interface (UI) screens. In
this paper, we present Ferret-UI, a new MLLM tailored for enhanced
understanding of mobile UI screens, equipped with referring, grounding, and
reasoning capabilities. Given that UI screens typically exhibit a more
elongated aspect ratio and contain smaller objects of interest (e.g., icons,
texts) than natural images, we incorporate "any resolution" on top of Ferret to
magnify details and leverage enhanced visual features. Specifically, each
screen is divided into 2 sub-images based on the original aspect ratio (i.e.,
horizontal division for portrait screens and vertical division for landscape
screens). Both sub-images are encoded separately before being sent to LLMs. We
meticulously gather training samples from an extensive range of elementary UI
tasks, such as icon recognition, find text, and widget listing. These samples
are formatted for instruction-following with region annotations to facilitate
precise referring and grounding. To augment the model's reasoning ability, we
further compile a dataset for advanced tasks, including detailed description,
perception/interaction conversations, and function inference. After training on
the curated datasets, Ferret-UI exhibits outstanding comprehension of UI
screens and the capability to execute open-ended instructions. For model
evaluation, we establish a comprehensive benchmark encompassing all the
aforementioned tasks. Ferret-UI excels not only beyond most open-source UI
MLLMs, but also surpasses GPT-4V on all the elementary UI tasks. | cs.CV | Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs |
2024-04-08T17:53:21Z | http://arxiv.org/pdf/2404.05783v1 | http://arxiv.org/abs/2404.05783v1 | Jindong Gu | 2024-04-08T17:53:21Z | cs.CY, cs.AI, cs.CL, cs.CV | In recent years, generative AI (GenAI), like large language models and
text-to-image models, has received significant attention across various
domains. However, ensuring the responsible generation of content by these
models is crucial for their real-world applicability. This raises an
interesting question: \textit{What should responsible GenAI generate, and what
should it not?} To answer the question, this paper investigates the practical
responsible requirements of both textual and visual generative models,
outlining five key considerations: generating truthful content, avoiding toxic
content, refusing harmful instruction, leaking no training data-related
content, and ensuring generated content identifiable. Specifically, we review
recent advancements and challenges in addressing these requirements. Besides,
we discuss and emphasize the importance of responsible GenAI across healthcare,
education, finance, and artificial general intelligence domains. Through a
unified perspective on both textual and visual generative models, this paper
aims to provide insights into practical safety-related issues and further
benefit the community in building responsible GenAI. | cs.CY | Responsible Generative AI: What to Generate and What Not |
2024-04-08T17:24:04Z | http://arxiv.org/pdf/2404.05694v2 | http://arxiv.org/abs/2404.05694v2 | Ahmad Idrissi-Yaghir, Amin Dada, Henning Schäfer, Kamyar Arzideh, Giulia Baldini, Jan Trienes, Max Hasin, Jeanette Bewersdorff, Cynthia S. Schmidt, Marie Bauer, Kaleb E. Smith, Jiang Bian, Yonghui Wu, Jörg Schlötterer, Torsten Zesch, Peter A. Horn, Christin Seifert, Felix Nensa, Jens Kleesiek, Christoph M. Friedrich | 2024-05-08T08:53:53Z | cs.CL, cs.AI, cs.LG | Recent advances in natural language processing (NLP) can be largely
attributed to the advent of pre-trained language models such as BERT and
RoBERTa. While these models demonstrate remarkable performance on general
datasets, they can struggle in specialized domains such as medicine, where
unique domain-specific terminologies, domain-specific abbreviations, and
varying document structures are common. This paper explores strategies for
adapting these models to domain-specific requirements, primarily through
continuous pre-training on domain-specific data. We pre-trained several German
medical language models on 2.4B tokens derived from translated public English
medical data and 3B tokens of German clinical data. The resulting models were
evaluated on various German downstream tasks, including named entity
recognition (NER), multi-label classification, and extractive question
answering. Our results suggest that models augmented by clinical and
translation-based pre-training typically outperform general domain models in
medical contexts. We conclude that continuous pre-training has demonstrated the
ability to match or even exceed the performance of clinical models trained from
scratch. Furthermore, pre-training on clinical data or leveraging translated
texts have proven to be reliable methods for domain adaptation in medical NLP
tasks. | cs.CL | Comprehensive Study on German Language Models for Clinical and Biomedical Text Understanding |
2024-04-08T17:18:04Z | http://arxiv.org/pdf/2404.05692v1 | http://arxiv.org/abs/2404.05692v1 | Shijie Xia, Xuefeng Li, Yixin Liu, Tongshuang Wu, Pengfei Liu | 2024-04-08T17:18:04Z | cs.CL | The leaderboard of Large Language Models (LLMs) in mathematical tasks has
been continuously updated. However, the majority of evaluations focus solely on
the final results, neglecting the quality of the intermediate steps. This
oversight can mask underlying problems, such as logical errors or unnecessary
steps in the reasoning process. To measure reasoning beyond final-answer
accuracy, we introduce ReasonEval, a new methodology for evaluating the quality
of reasoning steps. ReasonEval employs $\textit{validity}$ and
$\textit{redundancy}$ to characterize the reasoning quality, as well as
accompanying LLMs to assess them automatically. Instantiated by base models
that possess strong mathematical knowledge and trained with high-quality
labeled data, ReasonEval achieves state-of-the-art performance on human-labeled
datasets and can accurately detect different types of errors generated by
perturbation. When applied to evaluate LLMs specialized in math, we find that
an increase in final-answer accuracy does not necessarily guarantee an
improvement in the overall quality of the reasoning steps for challenging
mathematical problems. Additionally, we observe that ReasonEval can play a
significant role in data selection. We release the best-performing model,
meta-evaluation script, and all evaluation results at
https://github.com/GAIR-NLP/ReasonEval. | cs.CL | Evaluating Mathematical Reasoning Beyond Accuracy |
2024-04-08T16:43:52Z | http://arxiv.org/pdf/2404.05659v2 | http://arxiv.org/abs/2404.05659v2 | Khai Le-Duc | 2024-05-28T05:27:48Z | cs.CL, cs.AI, eess.AS | Due to privacy restrictions, there's a shortage of publicly available speech
recognition datasets in the medical domain. In this work, we present VietMed -
a Vietnamese speech recognition dataset in the medical domain comprising 16h of
labeled medical speech, 1000h of unlabeled medical speech and 1200h of
unlabeled general-domain speech. To our best knowledge, VietMed is by far the
world's largest public medical speech recognition dataset in 7 aspects: total
duration, number of speakers, diseases, recording conditions, speaker roles,
unique medical terms and accents. VietMed is also by far the largest public
Vietnamese speech dataset in terms of total duration. Additionally, we are the
first to present a medical ASR dataset covering all ICD-10 disease groups and
all accents within a country. Moreover, we release the first public large-scale
pre-trained models for Vietnamese ASR, w2v2-Viet and XLSR-53-Viet, along with
the first public large-scale fine-tuned models for medical ASR. Even without
any medical data in unsupervised pre-training, our best pre-trained model
XLSR-53-Viet generalizes very well to the medical domain by outperforming
state-of-the-art XLSR-53, from 51.8% to 29.6% WER on test set (a relative
reduction of more than 40%). All code, data and models are made publicly
available: https://github.com/leduckhai/MultiMed. | cs.CL | VietMed: A Dataset and Benchmark for Automatic Speech Recognition of Vietnamese in the Medical Domain |
2024-04-08T16:39:34Z | http://arxiv.org/pdf/2404.05656v2 | http://arxiv.org/abs/2404.05656v2 | Shahidur Rahoman Sohag, Sai Zhang, Min Xian, Shoukun Sun, Fei Xu, Zhegang Ma | 2024-04-22T15:25:20Z | cs.CL, cs.AI, cs.LG | Industry-wide nuclear power plant operating experience is a critical source
of raw data for performing parameter estimations in reliability and risk
models. Much operating experience information pertains to failure events and is
stored as reports containing unstructured data, such as narratives. Event
reports are essential for understanding how failures are initiated and
propagated, including the numerous causal relations involved. Causal relation
extraction using deep learning represents a significant frontier in the field
of natural language processing (NLP), and is crucial since it enables the
interpretation of intricate narratives and connections contained within vast
amounts of written information. This paper proposed a hybrid framework for
causality detection and extraction from nuclear licensee event reports. The
main contributions include: (1) we compiled an LER corpus with 20,129 text
samples for causality analysis, (2) developed an interactive tool for labeling
cause effect pairs, (3) built a deep-learning-based approach for causal
relation detection, and (4) developed a knowledge based cause-effect extraction
approach. | cs.CL | Causality Extraction from Nuclear Licensee Event Reports Using a Hybrid Framework |
2024-04-08T16:28:52Z | http://arxiv.org/pdf/2404.08684v1 | http://arxiv.org/abs/2404.08684v1 | Gian Alexandre Michaelsen, Renato P. dos Santos | 2024-04-08T16:28:52Z | cs.CL, cs.AI, cs.HC, cs.CY, J.4; K.3; I.2 | Background: The integration of artificial intelligence (AI) into daily life,
particularly through chatbots utilizing natural language processing (NLP),
presents both revolutionary potential and unique challenges. This intended to
investigate how different input forms impact ChatGPT, a leading language model
by OpenAI, performance in understanding and executing complex, multi-intention
tasks. Design: Employing a case study methodology supplemented by discourse
analysis, the research analyzes ChatGPT's responses to inputs varying from
natural language to pseudo-code engineering. The study specifically examines
the model's proficiency across four categories: understanding of intentions,
interpretability, completeness, and creativity. Setting and Participants: As a
theoretical exploration of AI interaction, this study focuses on the analysis
of structured and unstructured inputs processed by ChatGPT, without direct
human participants. Data collection and analysis: The research utilizes
synthetic case scenarios, including the organization of a "weekly meal plan"
and a "shopping list," to assess ChatGPT's response to prompts in both natural
language and pseudo-code engineering. The analysis is grounded in the
identification of patterns, contradictions, and unique response elements across
different input formats. Results: Findings reveal that pseudo-code engineering
inputs significantly enhance the clarity and determinism of ChatGPT's
responses, reducing ambiguity inherent in natural language. Enhanced natural
language, structured through prompt engineering techniques, similarly improves
the model's interpretability and creativity. Conclusions: The study underscores
the potential of pseudo-code engineering in refining human-AI interaction and
achieving more deterministic, concise, and direct outcomes, advocating for its
broader application across disciplines requiring precise AI responses. | cs.CL | Is English the New Programming Language? How About Pseudo-code Engineering? |
2024-04-08T16:18:33Z | http://arxiv.org/pdf/2404.08683v1 | http://arxiv.org/abs/2404.08683v1 | Lucas José Gonçalves Freitas, Thaís Rodrigues, Guilherme Rodrigues, Pamella Edokawa, Ariane Farias | 2024-04-08T16:18:33Z | cs.CL, cs.LG | Data analysis and machine learning are of preeminent importance in the legal
domain, especially in tasks like clustering and text classification. In this
study, we harnessed the power of natural language processing tools to enhance
datasets meticulously curated by experts. This process significantly improved
the classification workflow for legal texts using machine learning techniques.
We considered the Sustainable Development Goals (SDGs) data from the United
Nations 2030 Agenda as a practical case study. Data augmentation
clustering-based strategy led to remarkable enhancements in the accuracy and
sensitivity metrics of classification models. For certain SDGs within the 2030
Agenda, we observed performance gains of over 15%. In some cases, the example
base expanded by a noteworthy factor of 5. When dealing with unclassified legal
texts, data augmentation strategies centered around clustering prove to be
highly effective. They provide a valuable means to expand the existing
knowledge base without the need for labor-intensive manual classification
efforts. | cs.CL | Text clustering applied to data augmentation in legal contexts |
2024-04-08T16:04:26Z | http://arxiv.org/pdf/2404.05632v2 | http://arxiv.org/abs/2404.05632v2 | Haitham Hammami, Louis Baligand, Bojan Petrovski | 2024-04-09T09:30:46Z | cs.CL | In the financial industry, identifying the location of parties involved in
payments is a major challenge in the context of various regulatory
requirements. For this purpose address parsing entails extracting fields such
as street, postal code, or country from free text message attributes. While
payment processing platforms are updating their standards with more structured
formats such as SWIFT with ISO 20022, address parsing remains essential for a
considerable volume of messages. With the emergence of Transformers and
Generative Large Language Models (LLM), we explore the performance of
state-of-the-art solutions given the constraint of processing a vast amount of
daily data. This paper also aims to show the need for training robust models
capable of dealing with real-world noisy transactional data. Our results
suggest that a well fine-tuned Transformer model using early-stopping
significantly outperforms other approaches. Nevertheless, generative LLMs
demonstrate strong zero-shot performance and warrant further investigations. | cs.CL | Fighting crime with Transformers: Empirical analysis of address parsing methods in payment data |
2024-04-08T15:54:02Z | http://arxiv.org/pdf/2404.05624v1 | http://arxiv.org/abs/2404.05624v1 | Faren Yan, Peng Yu, Xin Chen | 2024-04-08T15:54:02Z | cs.CL, cs.AI | The use of LLMs for natural language processing has become a popular trend in
the past two years, driven by their formidable capacity for context
comprehension and learning, which has inspired a wave of research from
academics and industry professionals. However, for certain NLP tasks, such as
NER, the performance of LLMs still falls short when compared to supervised
learning methods. In our research, we developed a NER processing framework
called LTNER that incorporates a revolutionary Contextualized Entity Marking
Gen Method. By leveraging the cost-effective GPT-3.5 coupled with context
learning that does not require additional training, we significantly improved
the accuracy of LLMs in handling NER tasks. The F1 score on the CoNLL03 dataset
increased from the initial 85.9% to 91.9%, approaching the performance of
supervised fine-tuning. This outcome has led to a deeper understanding of the
potential of LLMs. | cs.CL | LTNER: Large Language Model Tagging for Named Entity Recognition with Contextualized Entity Marking |
2024-04-08T15:53:46Z | http://arxiv.org/pdf/2404.05623v2 | http://arxiv.org/abs/2404.05623v2 | Pietro Lesci, Andreas Vlachos | 2024-05-24T19:46:14Z | cs.LG, cs.CL | Active learning for imbalanced classification tasks is challenging as the
minority classes naturally occur rarely. Gathering a large pool of unlabelled
data is thus essential to capture minority instances. Standard pool-based
active learning is computationally expensive on large pools and often reaches
low accuracy by overfitting the initial decision boundary, thus failing to
explore the input space and find minority instances. To address these issues we
propose AnchorAL. At each iteration, AnchorAL chooses class-specific instances
from the labelled set, or anchors, and retrieves the most similar unlabelled
instances from the pool. This resulting subpool is then used for active
learning. Using a small, fixed-sized subpool AnchorAL allows scaling any active
learning strategy to large pools. By dynamically selecting different anchors at
each iteration it promotes class balance and prevents overfitting the initial
decision boundary, thus promoting the discovery of new clusters of minority
instances. In experiments across different classification tasks, active
learning strategies, and model architectures AnchorAL is (i) faster, often
reducing runtime from hours to minutes, (ii) trains more performant models,
(iii) and returns more balanced datasets than competing methods. | cs.LG | AnchorAL: Computationally Efficient Active Learning for Large and Imbalanced Datasets |
2024-04-08T15:53:29Z | http://arxiv.org/pdf/2404.05622v1 | http://arxiv.org/abs/2404.05622v1 | Olivier Binette, Youngsoo Baek, Siddharth Engineer, Christina Jones, Abel Dasylva, Jerome P. Reiter | 2024-04-08T15:53:29Z | cs.CL, cs.LG, stat.ME | Entity resolution (record linkage, microclustering) systems are notoriously
difficult to evaluate. Looking for a needle in a haystack, traditional
evaluation methods use sophisticated, application-specific sampling schemes to
find matching pairs of records among an immense number of non-matches. We
propose an alternative that facilitates the creation of representative,
reusable benchmark data sets without necessitating complex sampling schemes.
These benchmark data sets can then be used for model training and a variety of
evaluation tasks. Specifically, we propose an entity-centric data labeling
methodology that integrates with a unified framework for monitoring summary
statistics, estimating key performance metrics such as cluster and pairwise
precision and recall, and analyzing root causes for errors. We validate the
framework in an application to inventor name disambiguation and through
simulation studies. Software: https://github.com/OlivierBinette/er-evaluation/ | cs.CL | How to Evaluate Entity Resolution Systems: An Entity-Centric Framework with Application to Inventor Name Disambiguation |
2024-04-08T15:21:17Z | http://arxiv.org/pdf/2404.05600v1 | http://arxiv.org/abs/2404.05600v1 | Dong Zhang, Zhaowei Li, Shimin Li, Xin Zhang, Pengyu Wang, Yaqian Zhou, Xipeng Qiu | 2024-04-08T15:21:17Z | cs.CL, cs.SD, eess.AS | Speech language models have significantly advanced in generating realistic
speech, with neural codec language models standing out. However, the
integration of human feedback to align speech outputs to human preferences is
often neglected. This paper addresses this gap by first analyzing the
distribution gap in codec language models, highlighting how it leads to
discrepancies between the training and inference phases, which negatively
affects performance. Then we explore leveraging learning from human feedback to
bridge the distribution gap. We introduce SpeechAlign, an iterative
self-improvement strategy that aligns speech language models to human
preferences. SpeechAlign involves constructing a preference codec dataset
contrasting golden codec tokens against synthetic tokens, followed by
preference optimization to improve the codec language model. This cycle of
improvement is carried out iteratively to steadily convert weak models to
strong ones. Through both subjective and objective evaluations, we show that
SpeechAlign can bridge the distribution gap and facilitating continuous
self-improvement of the speech language model. Moreover, SpeechAlign exhibits
robust generalization capabilities and works for smaller models. Code and
models will be available at https://github.com/0nutation/SpeechGPT. | cs.CL | SpeechAlign: Aligning Speech Generation to Human Preferences |
2024-04-08T15:03:57Z | http://arxiv.org/pdf/2404.05590v1 | http://arxiv.org/abs/2404.05590v1 | Iñigo Alonso, Maite Oronoz, Rodrigo Agerri | 2024-04-08T15:03:57Z | cs.CL | Large Language Models (LLMs) have the potential of facilitating the
development of Artificial Intelligence technology to assist medical experts for
interactive decision support, which has been demonstrated by their competitive
performances in Medical QA. However, while impressive, the required quality bar
for medical applications remains far from being achieved. Currently, LLMs
remain challenged by outdated knowledge and by their tendency to generate
hallucinated content. Furthermore, most benchmarks to assess medical knowledge
lack reference gold explanations which means that it is not possible to
evaluate the reasoning of LLMs predictions. Finally, the situation is
particularly grim if we consider benchmarking LLMs for languages other than
English which remains, as far as we know, a totally neglected topic. In order
to address these shortcomings, in this paper we present MedExpQA, the first
multilingual benchmark based on medical exams to evaluate LLMs in Medical
Question Answering. To the best of our knowledge, MedExpQA includes for the
first time reference gold explanations written by medical doctors which can be
leveraged to establish various gold-based upper-bounds for comparison with LLMs
performance. Comprehensive multilingual experimentation using both the gold
reference explanations and Retrieval Augmented Generation (RAG) approaches show
that performance of LLMs still has large room for improvement, especially for
languages other than English. Furthermore, and despite using state-of-the-art
RAG methods, our results also demonstrate the difficulty of obtaining and
integrating readily available medical knowledge that may positively impact
results on downstream evaluations for Medical Question Answering. So far the
benchmark is available in four languages, but we hope that this work may
encourage further development to other languages. | cs.CL | MedExpQA: Multilingual Benchmarking of Large Language Models for Medical Question Answering |
2024-04-08T15:00:36Z | http://arxiv.org/pdf/2404.05587v2 | http://arxiv.org/abs/2404.05587v2 | Wolfgang Otto, Sharmila Upadhyaya, Stefan Dietze | 2024-04-19T23:19:17Z | cs.CL, I.2.7 | This paper describes our participation in the Shared Task on Software
Mentions Disambiguation (SOMD), with a focus on improving relation extraction
in scholarly texts through generative Large Language Models (LLMs) using
single-choice question-answering. The methodology prioritises the use of
in-context learning capabilities of GLMs to extract software-related entities
and their descriptive attributes, such as distributive information. Our
approach uses Retrieval-Augmented Generation (RAG) techniques and GLMs for
Named Entity Recognition (NER) and Attributive NER to identify relationships
between extracted software entities, providing a structured solution for
analysing software citations in academic literature. The paper provides a
detailed description of our approach, demonstrating how using GLMs in a
single-choice QA paradigm can greatly enhance IE methodologies. Our
participation in the SOMD shared task highlights the importance of precise
software citation practices and showcases our system's ability to overcome the
challenges of disambiguating and extracting relationships between software
mentions. This sets the groundwork for future research and development in this
field. | cs.CL | Enhancing Software-Related Information Extraction via Single-Choice Question Answering with Large Language Models |
2024-04-08T14:43:13Z | http://arxiv.org/pdf/2404.05569v2 | http://arxiv.org/abs/2404.05569v2 | Shen Gao, Hao Li, Chengrui Huang, Quan Tu, Zhiliang Tian, Minlie Huang, Shuo Shang | 2024-06-26T11:42:10Z | cs.AI, cs.CL, cs.MA | Large language model agents have demonstrated remarkable advancements across
various complex tasks. Recent works focus on optimizing the agent team or
employing self-reflection to iteratively solve complex tasks. Since these
agents are all based on the same LLM, only conducting self-evaluation or
removing underperforming agents does not substantively enhance the capability
of the agents. We argue that a comprehensive evaluation and accumulating
experience from evaluation feedback is an effective approach to improving
system performance. In this paper, we propose Reusable Experience Accumulation
with 360$^\circ$ Assessment (360$^\circ$REA), a hierarchical multi-agent
framework inspired by corporate organizational practices. The framework employs
a novel 360$^\circ$ performance assessment method for multi-perspective
performance evaluation with fine-grained assessment. To enhance the capability
of agents in addressing complex tasks, we introduce dual-level experience pool
for agents to accumulate experience through fine-grained assessment. Extensive
experiments on complex task datasets demonstrate the effectiveness of
360$^\circ$REA. | cs.AI | 360$^\circ$REA: Towards A Reusable Experience Accumulation with 360° Assessment for Multi-Agent System |
2024-04-08T14:39:49Z | http://arxiv.org/pdf/2404.05567v1 | http://arxiv.org/abs/2404.05567v1 | Bowen Pan, Yikang Shen, Haokun Liu, Mayank Mishra, Gaoyuan Zhang, Aude Oliva, Colin Raffel, Rameswar Panda | 2024-04-08T14:39:49Z | cs.LG, cs.AI, cs.CL | Mixture-of-Experts (MoE) language models can reduce computational costs by
2-4$\times$ compared to dense models without sacrificing performance, making
them more efficient in computation-bounded scenarios. However, MoE models
generally require 2-4$\times$ times more parameters to achieve comparable
performance to a dense model, which incurs larger GPU memory requirements and
makes MoE models less efficient in I/O-bounded scenarios like autoregressive
generation. In this work, we propose a hybrid dense training and sparse
inference framework for MoE models (DS-MoE) which achieves strong computation
and parameter efficiency by employing dense computation across all experts
during training and sparse computation during inference. Our experiments on
training LLMs demonstrate that our DS-MoE models are more parameter-efficient
than standard sparse MoEs and are on par with dense models in terms of total
parameter size and performance while being computationally cheaper (activating
30-40% of the model's parameters). Performance tests using vLLM show that our
DS-MoE-6B model runs up to $1.86\times$ faster than similar dense models like
Mistral-7B, and between $1.50\times$ and $1.71\times$ faster than comparable
MoEs, such as DeepSeekMoE-16B and Qwen1.5-MoE-A2.7B. | cs.LG | Dense Training, Sparse Inference: Rethinking Training of Mixture-of-Experts Language Models |
2024-04-08T14:32:52Z | http://arxiv.org/pdf/2404.05560v1 | http://arxiv.org/abs/2404.05560v1 | Longhui Zhang, Dingkun Long, Meishan Zhang, Yanzhao Zhang, Pengjun Xie, Min Zhang | 2024-04-08T14:32:52Z | cs.CL | Chinese sequence labeling tasks are heavily reliant on accurate word boundary
demarcation. Although current pre-trained language models (PLMs) have achieved
substantial gains on these tasks, they rarely explicitly incorporate boundary
information into the modeling process. An exception to this is BABERT, which
incorporates unsupervised statistical boundary information into Chinese BERT's
pre-training objectives. Building upon this approach, we input supervised
high-quality boundary information to enhance BABERT's learning, developing a
semi-supervised boundary-aware PLM. To assess PLMs' ability to encode
boundaries, we introduce a novel ``Boundary Information Metric'' that is both
simple and effective. This metric allows comparison of different PLMs without
task-specific fine-tuning. Experimental results on Chinese sequence labeling
datasets demonstrate that the improved BABERT variant outperforms the vanilla
version, not only on these tasks but also more broadly across a range of
Chinese natural language understanding tasks. Additionally, our proposed metric
offers a convenient and accurate means of evaluating PLMs' boundary awareness. | cs.CL | Chinese Sequence Labeling with Semi-Supervised Boundary-Aware Language Model Pre-training |
2024-04-08T14:15:56Z | http://arxiv.org/pdf/2404.05545v1 | http://arxiv.org/abs/2404.05545v1 | Tejas Kasetty, Divyat Mahajan, Gintare Karolina Dziugaite, Alexandre Drouin, Dhanya Sridhar | 2024-04-08T14:15:56Z | cs.LG, cs.AI, cs.CL, stat.ME | Numerous decision-making tasks require estimating causal effects under
interventions on different parts of a system. As practitioners consider using
large language models (LLMs) to automate decisions, studying their causal
reasoning capabilities becomes crucial. A recent line of work evaluates LLMs
ability to retrieve commonsense causal facts, but these evaluations do not
sufficiently assess how LLMs reason about interventions. Motivated by the role
that interventions play in causal inference, in this paper, we conduct
empirical analyses to evaluate whether LLMs can accurately update their
knowledge of a data-generating process in response to an intervention. We
create benchmarks that span diverse causal graphs (e.g., confounding,
mediation) and variable types, and enable a study of intervention-based
reasoning. These benchmarks allow us to isolate the ability of LLMs to
accurately predict changes resulting from their ability to memorize facts or
find other shortcuts. Our analysis on four LLMs highlights that while GPT- 4
models show promising accuracy at predicting the intervention effects, they
remain sensitive to distracting factors in the prompts. | cs.LG | Evaluating Interventional Reasoning Capabilities of Large Language Models |
2024-04-08T14:08:56Z | http://arxiv.org/pdf/2404.05540v1 | http://arxiv.org/abs/2404.05540v1 | Mehran Safayani, Amir Sartipi, Amir Hossein Ahmadi, Parniyan Jalali, Amir Hossein Mansouri, Mohammad Bisheh-Niasar, Zahra Pourbahman | 2024-04-08T14:08:56Z | cs.CL, cs.AI, cs.LG | The proliferation of hate speech and offensive comments on social media has
become increasingly prevalent due to user activities. Such comments can have
detrimental effects on individuals' psychological well-being and social
behavior. While numerous datasets in the English language exist in this domain,
few equivalent resources are available for Persian language. To address this
gap, this paper introduces two offensive datasets. The first dataset comprises
annotations provided by domain experts, while the second consists of a large
collection of unlabeled data obtained through web crawling for unsupervised
learning purposes. To ensure the quality of the former dataset, a meticulous
three-stage labeling process was conducted, and kappa measures were computed to
assess inter-annotator agreement. Furthermore, experiments were performed on
the dataset using state-of-the-art language models, both with and without
employing masked language modeling techniques, as well as machine learning
algorithms, in order to establish the baselines for the dataset using
contemporary cutting-edge approaches. The obtained F1-scores for the
three-class and two-class versions of the dataset were 76.9% and 89.9% for
XLM-RoBERTa, respectively. | cs.CL | OPSD: an Offensive Persian Social media Dataset and its baseline evaluations |
2024-04-08T13:59:02Z | http://arxiv.org/pdf/2404.05530v1 | http://arxiv.org/abs/2404.05530v1 | Tim Baumgärtner, Yang Gao, Dana Alon, Donald Metzler | 2024-04-08T13:59:02Z | cs.CL, cs.AI, cs.CR, cs.LG | Reinforcement Learning from Human Feedback (RLHF) is a popular method for
aligning Language Models (LM) with human values and preferences. RLHF requires
a large number of preference pairs as training data, which are often used in
both the Supervised Fine-Tuning and Reward Model training, and therefore
publicly available datasets are commonly used. In this work, we study to what
extent a malicious actor can manipulate the LMs generations by poisoning the
preferences, i.e., injecting poisonous preference pairs into these datasets and
the RLHF training process. We propose strategies to build poisonous preference
pairs and test their performance by poisoning two widely used preference
datasets. Our results show that preference poisoning is highly effective: by
injecting a small amount of poisonous data (1-5% of the original dataset), we
can effectively manipulate the LM to generate a target entity in a target
sentiment (positive or negative). The findings from our experiments also shed
light on strategies to defend against the preference poisoning attack. | cs.CL | Best-of-Venom: Attacking RLHF by Injecting Poisoned Preference Data |
2024-04-08T13:28:11Z | http://arxiv.org/pdf/2404.05508v1 | http://arxiv.org/abs/2404.05508v1 | Nenad Petrovic, Fengjunjie Pan, Krzysztof Lebioda, Vahid Zolfaghari, Sven Kirchner, Nils Purschke, Muhammad Aqib Khan, Viktor Vorobev, Alois Knoll | 2024-04-08T13:28:11Z | cs.SE, cs.AI, cs.CL, D.2.1; D.2.2; D.2.4; I.2.7; I.2.2; I.7.0 | We present a prototype of a tool leveraging the synergy of model driven
engineering (MDE) and Large Language Models (LLM) for the purpose of software
development process automation in the automotive industry. In this approach,
the user-provided input is free form textual requirements, which are first
translated to Ecore model instance representation using an LLM, which is
afterwards checked for consistency using Object Constraint Language (OCL)
rules. After successful consistency check, the model instance is fed as input
to another LLM for the purpose of code generation. The generated code is
evaluated in a simulated environment using CARLA simulator connected to an
example centralized vehicle architecture, in an emergency brake scenario. | cs.SE | Synergy of Large Language Model and Model Driven Engineering for Automated Development of Centralized Vehicular Systems |
2024-04-08T13:25:03Z | http://arxiv.org/pdf/2404.05502v1 | http://arxiv.org/abs/2404.05502v1 | Roman Kazakov, Kseniia Petukhova, Ekaterina Kochmar | 2024-04-08T13:25:03Z | cs.CL, cs.AI, I.2.7 | In this paper, we present our submission to the SemEval-2023 Task~3 "The
Competition of Multimodal Emotion Cause Analysis in Conversations", focusing on
extracting emotion-cause pairs from dialogs. Specifically, our approach relies
on combining fine-tuned GPT-3.5 for emotion classification and a BiLSTM-based
neural network to detect causes. We score 2nd in the ranking for Subtask 1,
demonstrating the effectiveness of our approach through one of the highest
weighted-average proportional F1 scores recorded at 0.264. | cs.CL | PetKaz at SemEval-2024 Task 3: Advancing Emotion Classification with an LLM for Emotion-Cause Pair Extraction in Conversations |
2024-04-08T13:05:02Z | http://arxiv.org/pdf/2404.05483v1 | http://arxiv.org/abs/2404.05483v1 | Kseniia Petukhova, Roman Kazakov, Ekaterina Kochmar | 2024-04-08T13:05:02Z | cs.CL, cs.AI, I.2.7 | In this paper, we present our submission to the SemEval-2024 Task 8
"Multigenerator, Multidomain, and Multilingual Black-Box Machine-Generated Text
Detection", focusing on the detection of machine-generated texts (MGTs) in
English. Specifically, our approach relies on combining embeddings from the
RoBERTa-base with diversity features and uses a resampled training set. We
score 12th from 124 in the ranking for Subtask A (monolingual track), and our
results show that our approach is generalizable across unseen models and
domains, achieving an accuracy of 0.91. | cs.CL | PetKaz at SemEval-2024 Task 8: Can Linguistics Capture the Specifics of LLM-generated Text? |
2024-04-08T12:31:23Z | http://arxiv.org/pdf/2404.05449v2 | http://arxiv.org/abs/2404.05449v2 | Wenyang Hui, Chengyue Jiang, Yan Wang, Kewei Tu | 2024-04-11T05:21:00Z | cs.CL | Large language models (LLMs) have demonstrated impressive capability in
reasoning and planning when integrated with tree-search-based prompting
methods. However, since these methods ignore the previous search experiences,
they often make the same mistakes in the search process. To address this issue,
we introduce Reflection on search Trees (RoT), an LLM reflection framework
designed to improve the performance of tree-search-based prompting methods. It
uses a strong LLM to summarize guidelines from previous tree search experiences
to enhance the ability of a weak LLM. The guidelines are instructions about
solving this task through tree search which can prevent the weak LLMs from
making similar mistakes in the past search process. In addition, we proposed a
novel state selection method, which identifies the critical information from
historical search processes to help RoT generate more specific and meaningful
guidelines. In our extensive experiments, we find that RoT significantly
improves the performance of LLMs in reasoning or planning tasks with various
tree-search-based prompting methods (e.g., BFS and MCTS). Non-tree-search-based
prompting methods such as Chain-of-Thought (CoT) can also benefit from RoT
guidelines since RoT can provide task-specific knowledge collected from the
search experience. | cs.CL | RoT: Enhancing Large Language Models with Reflection on Search Trees |
2024-04-08T12:29:07Z | http://arxiv.org/pdf/2404.05446v1 | http://arxiv.org/abs/2404.05446v1 | Xuanfan Ni, Hengyi Cai, Xiaochi Wei, Shuaiqiang Wang, Dawei Yin, Piji Li | 2024-04-08T12:29:07Z | cs.CL | Large Language Models (LLMs) have demonstrated remarkable performance across
diverse tasks but are constrained by their small context window sizes. Various
efforts have been proposed to expand the context window to accommodate even up
to 200K input tokens. Meanwhile, building high-quality benchmarks with much
longer text lengths and more demanding tasks to provide comprehensive
evaluations is of immense practical interest to facilitate long context
understanding research of LLMs. However, prior benchmarks create datasets that
ostensibly cater to long-text comprehension by expanding the input of
traditional tasks, which falls short to exhibit the unique characteristics of
long-text understanding, including long dependency tasks and longer text length
compatible with modern LLMs' context window size. In this paper, we introduce a
benchmark for extremely long context understanding with long-range
dependencies, XL$^2$Bench, which includes three scenarios: Fiction Reading,
Paper Reading, and Law Reading, and four tasks of increasing complexity: Memory
Retrieval, Detailed Understanding, Overall Understanding, and Open-ended
Generation, covering 27 subtasks in English and Chinese. It has an average
length of 100K+ words (English) and 200K+ characters (Chinese). Evaluating six
leading LLMs on XL$^2$Bench, we find that their performance significantly lags
behind human levels. Moreover, the observed decline in performance across both
the original and enhanced datasets underscores the efficacy of our approach to
mitigating data contamination. | cs.CL | XL$^2$Bench: A Benchmark for Extremely Long Context Understanding with Long-range Dependencies |
2024-04-08T11:55:44Z | http://arxiv.org/pdf/2404.05428v1 | http://arxiv.org/abs/2404.05428v1 | Nikola Ljubešić, Vít Suchomel, Peter Rupnik, Taja Kuzman, Rik van Noord | 2024-04-08T11:55:44Z | cs.CL | The world of language models is going through turbulent times, better and
ever larger models are coming out at an unprecedented speed. However, we argue
that, especially for the scientific community, encoder models of up to 1
billion parameters are still very much needed, their primary usage being in
enriching large collections of data with metadata necessary for downstream
research. We investigate the best way to ensure the existence of such encoder
models on the set of very closely related languages - Croatian, Serbian,
Bosnian and Montenegrin, by setting up a diverse benchmark for these languages,
and comparing the trained-from-scratch models with the new models constructed
via additional pretraining of existing multilingual models. We show that
comparable performance to dedicated from-scratch models can be obtained by
additionally pretraining available multilingual models even with a limited
amount of computation. We also show that neighboring languages, in our case
Slovenian, can be included in the additional pretraining with little to no loss
in the performance of the final model. | cs.CL | Language Models on a Diet: Cost-Efficient Development of Encoders for Closely-Related Languages via Additional Pretraining |
2024-04-08T11:33:00Z | http://arxiv.org/pdf/2404.05415v2 | http://arxiv.org/abs/2404.05415v2 | Yiming Li, Xueqing Peng, Jianfu Li, Xu Zuo, Suyuan Peng, Donghong Pei, Cui Tao, Hua Xu, Na Hong | 2024-04-15T00:45:32Z | cs.CL, cs.AI | In acupuncture therapy, the accurate location of acupoints is essential for
its effectiveness. The advanced language understanding capabilities of large
language models (LLMs) like Generative Pre-trained Transformers (GPT) present a
significant opportunity for extracting relations related to acupoint locations
from textual knowledge sources. This study aims to compare the performance of
GPT with traditional deep learning models (Long Short-Term Memory (LSTM) and
Bidirectional Encoder Representations from Transformers for Biomedical Text
Mining (BioBERT)) in extracting acupoint-related location relations and assess
the impact of pretraining and fine-tuning on GPT's performance. We utilized the
World Health Organization Standard Acupuncture Point Locations in the Western
Pacific Region (WHO Standard) as our corpus, which consists of descriptions of
361 acupoints. Five types of relations ('direction_of,' 'distance_of,'
'part_of,' 'near_acupoint,' and 'located_near') (n= 3,174) between acupoints
were annotated. Five models were compared: BioBERT, LSTM, pre-trained GPT-3.5,
fine-tuned GPT-3.5, as well as pre-trained GPT-4. Performance metrics included
micro-average exact match precision, recall, and F1 scores. Our results
demonstrate that fine-tuned GPT-3.5 consistently outperformed other models in
F1 scores across all relation types. Overall, it achieved the highest
micro-average F1 score of 0.92. This study underscores the effectiveness of
LLMs like GPT in extracting relations related to acupoint locations, with
implications for accurately modeling acupuncture knowledge and promoting
standard implementation in acupuncture training and practice. The findings also
contribute to advancing informatics applications in traditional and
complementary medicine, showcasing the potential of LLMs in natural language
processing. | cs.CL | Relation Extraction Using Large Language Models: A Case Study on Acupuncture Point Locations |
2024-04-08T11:25:30Z | http://arxiv.org/pdf/2404.05411v1 | http://arxiv.org/abs/2404.05411v1 | Ava Spataru, Eric Hambro, Elena Voita, Nicola Cancedda | 2024-04-08T11:25:30Z | cs.CL | In this work, we explicitly show that modern LLMs tend to generate correct
facts first, then "drift away" and generate incorrect facts later: this was
occasionally observed but never properly measured. We develop a semantic drift
score that measures the degree of separation between correct and incorrect
facts in generated texts and confirm our hypothesis when generating
Wikipedia-style biographies. This correct-then-incorrect generation pattern
suggests that factual accuracy can be improved by knowing when to stop
generation. Therefore, we explore the trade-off between information quantity
and factual accuracy for several early stopping methods and manage to improve
factuality by a large margin. We further show that reranking with semantic
similarity can further improve these results, both compared to the baseline and
when combined with early stopping. Finally, we try calling external API to
bring the model back to the right generation path, but do not get positive
results. Overall, our methods generalize and can be applied to any long-form
text generation to produce more reliable information, by balancing trade-offs
between factual accuracy, information quantity and computational cost. | cs.CL | Know When To Stop: A Study of Semantic Drift in Text Generation |
2024-04-08T11:14:58Z | http://arxiv.org/pdf/2404.05406v2 | http://arxiv.org/abs/2404.05406v2 | Pardis Moradbeiki, Nasser Ghadiri | 2024-04-15T12:38:33Z | cs.CL, cs.AI, 68T50, 68T07, J.3; I.2.7; I.2.1 | Smart cities need the involvement of their residents to enhance quality of
life. Conversational query-answering is an emerging approach for user
engagement. There is an increasing demand of an advanced conversational
question-answering that goes beyond classic systems. Existing approaches have
shown that LLMs offer promising capabilities for CQA, but may struggle to
capture the nuances of conversational contexts. The new approach involves
understanding the content and engaging in a multi-step conversation with the
user to fulfill their needs. This paper presents a novel method to elevate the
performance of Persian Conversational question-answering (CQA) systems. It
combines the strengths of Large Language Models (LLMs) with contextual keyword
extraction. Our method extracts keywords specific to the conversational flow,
providing the LLM with additional context to understand the user's intent and
generate more relevant and coherent responses. We evaluated the effectiveness
of this combined approach through various metrics, demonstrating significant
improvements in CQA performance compared to an LLM-only baseline. The proposed
method effectively handles implicit questions, delivers contextually relevant
answers, and tackles complex questions that rely heavily on conversational
context. The findings indicate that our method outperformed the evaluation
benchmarks up to 8% higher than existing methods and the LLM-only baseline. | cs.CL | PerkwE_COQA: Enhanced Persian Conversational Question Answering by combining contextual keyword extraction with Large Language Models |
2024-04-08T11:11:31Z | http://arxiv.org/pdf/2404.05405v1 | http://arxiv.org/abs/2404.05405v1 | Zeyuan Allen-Zhu, Yuanzhi Li | 2024-04-08T11:11:31Z | cs.CL, cs.AI, cs.LG | Scaling laws describe the relationship between the size of language models
and their capabilities. Unlike prior studies that evaluate a model's capability
via loss or benchmarks, we estimate the number of knowledge bits a model
stores. We focus on factual knowledge represented as tuples, such as (USA,
capital, Washington D.C.) from a Wikipedia page. Through multiple controlled
datasets, we establish that language models can and only can store 2 bits of
knowledge per parameter, even when quantized to int8, and such knowledge can be
flexibly extracted for downstream applications. Consequently, a 7B model can
store 14B bits of knowledge, surpassing the English Wikipedia and textbooks
combined based on our estimation.
More broadly, we present 12 results on how (1) training duration, (2) model
architecture, (3) quantization, (4) sparsity constraints such as MoE, and (5)
data signal-to-noise ratio affect a model's knowledge storage capacity. Notable
insights include:
* The GPT-2 architecture, with rotary embedding, matches or even surpasses
LLaMA/Mistral architectures in knowledge storage, particularly over shorter
training durations. This arises because LLaMA/Mistral uses GatedMLP, which is
less stable and harder to train.
* Prepending training data with domain names (e.g., wikipedia.org)
significantly increases a model's knowledge capacity. Language models can
autonomously identify and prioritize domains rich in knowledge, optimizing
their storage capacity. | cs.CL | Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws |
2024-04-08T10:57:25Z | http://arxiv.org/pdf/2404.05399v1 | http://arxiv.org/abs/2404.05399v1 | Paul Röttger, Fabio Pernisi, Bertie Vidgen, Dirk Hovy | 2024-04-08T10:57:25Z | cs.CL, cs.AI | The last two years have seen a rapid growth in concerns around the safety of
large language models (LLMs). Researchers and practitioners have met these
concerns by introducing an abundance of new datasets for evaluating and
improving LLM safety. However, much of this work has happened in parallel, and
with very different goals in mind, ranging from the mitigation of near-term
risks around bias and toxic content generation to the assessment of longer-term
catastrophic risk potential. This makes it difficult for researchers and
practitioners to find the most relevant datasets for a given use case, and to
identify gaps in dataset coverage that future work may fill. To remedy these
issues, we conduct a first systematic review of open datasets for evaluating
and improving LLM safety. We review 102 datasets, which we identified through
an iterative and community-driven process over the course of several months. We
highlight patterns and trends, such as a a trend towards fully synthetic
datasets, as well as gaps in dataset coverage, such as a clear lack of
non-English datasets. We also examine how LLM safety datasets are used in
practice -- in LLM release publications and popular LLM benchmarks -- finding
that current evaluation practices are highly idiosyncratic and make use of only
a small fraction of available datasets. Our contributions are based on
SafetyPrompts.com, a living catalogue of open datasets for LLM safety, which we
commit to updating continuously as the field of LLM safety develops. | cs.CL | SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety |
2024-04-08T10:04:55Z | http://arxiv.org/pdf/2404.05365v2 | http://arxiv.org/abs/2404.05365v2 | Atnafu Lambebo Tonja, Fazlourrahman Balouchzahi, Sabur Butt, Olga Kolesnikova, Hector Ceballos, Alexander Gelbukh, Thamar Solorio | 2024-05-12T13:39:42Z | cs.CL | The paper focuses on the marginalization of indigenous language communities
in the face of rapid technological advancements. We highlight the cultural
richness of these languages and the risk they face of being overlooked in the
realm of Natural Language Processing (NLP). We aim to bridge the gap between
these communities and researchers, emphasizing the need for inclusive
technological advancements that respect indigenous community perspectives. We
show the NLP progress of indigenous Latin American languages and the survey
that covers the status of indigenous languages in Latin America, their
representation in NLP, and the challenges and innovations required for their
preservation and development. The paper contributes to the current literature
in understanding the need and progress of NLP for indigenous communities of
Latin America, specifically low-resource and indigenous communities in general. | cs.CL | NLP Progress in Indigenous Latin American Languages |
2024-04-08T09:25:32Z | http://arxiv.org/pdf/2404.05337v1 | http://arxiv.org/abs/2404.05337v1 | Chenxu Wang, Bin Dai, Huaping Liu, Baoyuan Wang | 2024-04-08T09:25:32Z | cs.CL, cs.AI | Prominent large language models have exhibited human-level performance in
many domains, even enabling the derived agents to simulate human and social
interactions. While practical works have substantiated the practicability of
grounding language agents in sandbox simulation or embodied simulators, current
social intelligence benchmarks either stay at the language level or use
subjective metrics. In pursuit of a more realistic and objective evaluation, we
introduce the Social Tasks in Sandbox Simulation (STSS) benchmark, which
assesses language agents \textbf{objectively} at the \textbf{action level} by
scrutinizing the goal achievements within the multi-agent simulation.
Additionally, we sample conversation scenarios to build a language-level
benchmark to provide an economically prudent preliminary evaluation and align
with prevailing benchmarks. To gauge the significance of agent architecture, we
implement a target-driven planning (TDP) module as an adjunct to the existing
agent. Our evaluative findings highlight that the STSS benchmark is challenging
for state-of-the-art language agents. Furthermore, it effectively discriminates
between distinct language agents, suggesting its usefulness as a benchmark for
evaluating both language models and agent architectures. | cs.CL | Towards Objectively Benchmarking Social Intelligence for Language Agents at Action Level |
2024-04-08T09:22:41Z | http://arxiv.org/pdf/2404.05333v3 | http://arxiv.org/abs/2404.05333v3 | Tomás Osório, Bernardo Leite, Henrique Lopes Cardoso, Luís Gomes, João Rodrigues, Rodrigo Santos, António Branco | 2024-05-08T19:32:42Z | cs.CL | Leveraging research on the neural modelling of Portuguese, we contribute a
collection of datasets for an array of language processing tasks and a
corresponding collection of fine-tuned neural language models on these
downstream tasks. To align with mainstream benchmarks in the literature,
originally developed in English, and to kick start their Portuguese
counterparts, the datasets were machine-translated from English with a
state-of-the-art translation engine. The resulting PORTULAN ExtraGLUE benchmark
is a basis for research on Portuguese whose improvement can be pursued in
future work. Similarly, the respective fine-tuned neural language models,
developed with a low-rank adaptation approach, are made available as baselines
that can stimulate future work on the neural processing of Portuguese. All
datasets and models have been developed and are made available for two variants
of Portuguese: European and Brazilian. | cs.CL | PORTULAN ExtraGLUE Datasets and Models: Kick-starting a Benchmark for the Neural Processing of Portuguese |
2024-04-08T08:13:40Z | http://arxiv.org/pdf/2404.05281v1 | http://arxiv.org/abs/2404.05281v1 | Syrielle Montariol, Matej Martinc, Andraž Pelicon, Senja Pollak, Boshko Koloski, Igor Lončarski, Aljoša Valentinčič | 2024-04-08T08:13:40Z | cs.CL | For assessing various performance indicators of companies, the focus is
shifting from strictly financial (quantitative) publicly disclosed information
to qualitative (textual) information. This textual data can provide valuable
weak signals, for example through stylistic features, which can complement the
quantitative data on financial performance or on Environmental, Social and
Governance (ESG) criteria. In this work, we use various multi-task learning
methods for financial text classification with the focus on financial
sentiment, objectivity, forward-looking sentence prediction and ESG-content
detection. We propose different methods to combine the information extracted
from training jointly on different tasks; our best-performing method highlights
the positive effect of explicitly adding auxiliary task predictions as features
for the final target task during the multi-task training. Next, we use these
classifiers to extract textual features from annual reports of FTSE350
companies and investigate the link between ESG quantitative scores and these
features. | cs.CL | Multi-Task Learning for Features Extraction in Financial Annual Reports |
2024-04-08T07:37:31Z | http://arxiv.org/pdf/2404.08001v1 | http://arxiv.org/abs/2404.08001v1 | Zhengde Zhang, Yiyu Zhang, Haodong Yao, Jianwen Luo, Rui Zhao, Bo Huang, Jiameng Zhao, Yipu Liao, Ke Li, Lina Zhao, Jun Cao, Fazhi Qi, Changzheng Yuan | 2024-04-08T07:37:31Z | hep-ph, cs.AI, cs.CL, cs.LG, hep-ex, physics.comp-ph, I.2.7 | Large Language Models (LLMs) are undergoing a period of rapid updates and
changes, with state-of-the-art (SOTA) model frequently being replaced. When
applying LLMs to a specific scientific field, it's challenging to acquire
unique domain knowledge while keeping the model itself advanced. To address
this challenge, a sophisticated large language model system named as Xiwu has
been developed, allowing you switch between the most advanced foundation models
and quickly teach the model domain knowledge. In this work, we will report on
the best practices for applying LLMs in the field of high-energy physics (HEP),
including: a seed fission technology is proposed and some data collection and
cleaning tools are developed to quickly obtain domain AI-Ready dataset; a
just-in-time learning system is implemented based on the vector store
technology; an on-the-fly fine-tuning system has been developed to facilitate
rapid training under a specified foundation model. The results show that Xiwu
can smoothly switch between foundation models such as LLaMA, Vicuna, ChatGLM
and Grok-1. The trained Xiwu model is significantly outperformed the benchmark
model on the HEP knowledge question-and-answering and code generation. This
strategy significantly enhances the potential for growth of our model's
performance, with the hope of surpassing GPT-4 as it evolves with the
development of open-source models. This work provides a customized LLM for the
field of HEP, while also offering references for applying LLM to other fields,
the corresponding codes are available on Github. | hep-ph | Xiwu: A Basis Flexible and Learnable LLM for High Energy Physics |
2024-04-08T07:36:26Z | http://arxiv.org/pdf/2404.08681v1 | http://arxiv.org/abs/2404.08681v1 | Tianyu Chen, Yiming Zhang, Guoxin Yu, Dapeng Zhang, Li Zeng, Qing He, Xiang Ao | 2024-04-08T07:36:26Z | cs.CL | In this paper, we extend financial sentiment analysis~(FSA) to event-level
since events usually serve as the subject of the sentiment in financial text.
Though extracting events from the financial text may be conducive to accurate
sentiment predictions, it has specialized challenges due to the lengthy and
discontinuity of events in a financial text. To this end, we reconceptualize
the event extraction as a classification task by designing a categorization
comprising coarse-grained and fine-grained event categories. Under this
setting, we formulate the \textbf{E}vent-Level \textbf{F}inancial
\textbf{S}entiment \textbf{A}nalysis~(\textbf{EFSA} for short) task that
outputs quintuples consisting of (company, industry, coarse-grained event,
fine-grained event, sentiment) from financial text. A large-scale Chinese
dataset containing $12,160$ news articles and $13,725$ quintuples is publicized
as a brand new testbed for our task. A four-hop Chain-of-Thought LLM-based
approach is devised for this task. Systematically investigations are conducted
on our dataset, and the empirical results demonstrate the benchmarking scores
of existing methods and our proposed method can reach the current
state-of-the-art. Our dataset and framework implementation are available at
https://anonymous.4open.science/r/EFSA-645E | cs.CL | EFSA: Towards Event-Level Financial Sentiment Analysis |
2024-04-08T07:26:27Z | http://arxiv.org/pdf/2404.05250v1 | http://arxiv.org/abs/2404.05250v1 | Yigeng Zhang, Fabio A. González, Thamar Solorio | 2024-04-08T07:26:27Z | cs.CL | Reading comprehension continues to be a crucial research focus in the NLP
community. Recent advances in Machine Reading Comprehension (MRC) have mostly
centered on literal comprehension, referring to the surface-level understanding
of content. In this work, we focus on the next level - interpretive
comprehension, with a particular emphasis on inferring the themes of a
narrative text. We introduce the first dataset specifically designed for
interpretive comprehension of educational narratives, providing corresponding
well-edited theme texts. The dataset spans a variety of genres and cultural
origins and includes human-annotated theme keywords with varying levels of
granularity. We further formulate NLP tasks under different abstractions of
interpretive comprehension toward the main idea of a story. After conducting
extensive experiments with state-of-the-art methods, we found the task to be
both challenging and significant for NLP research. The dataset and source code
have been made publicly available to the research community at
https://github.com/RiTUAL-UH/EduStory. | cs.CL | Interpreting Themes from Educational Stories |
2024-04-08T07:21:46Z | http://arxiv.org/pdf/2404.05245v1 | http://arxiv.org/abs/2404.05245v1 | Murtadha Ahmed, Qun Chen | 2024-04-08T07:21:46Z | cs.CL | Aspect Category Detection (ACD) aims to identify implicit and explicit
aspects in a given review sentence. The state-of-the-art approaches for ACD use
Deep Neural Networks (DNNs) to address the problem as a multi-label
classification task. However, learning category-specific representations
heavily rely on the amount of labeled examples, which may not readily available
in real-world scenarios. In this paper, we propose a novel approach to tackle
the ACD task by combining DNNs with Gradual Machine Learning (GML) in a
supervised setting. we aim to leverage the strength of DNN in semantic relation
modeling, which can facilitate effective knowledge transfer between labeled and
unlabeled instances during the gradual inference of GML. To achieve this, we
first analyze the learned latent space of the DNN to model the relations, i.e.,
similar or opposite, between instances. We then represent these relations as
binary features in a factor graph to efficiently convey knowledge. Finally, we
conduct a comparative study of our proposed solution on real benchmark datasets
and demonstrate that the GML approach, in collaboration with DNNs for feature
extraction, consistently outperforms pure DNN solutions. | cs.CL | Supervised Gradual Machine Learning for Aspect Category Detection |
2024-04-08T07:15:06Z | http://arxiv.org/pdf/2404.05243v1 | http://arxiv.org/abs/2404.05243v1 | Tejpalsingh Siledar, Rupasai Rangaraju, Sankara Sri Raghava Ravindra Muddu, Suman Banerjee, Amey Patil, Sudhanshu Shekhar Singh, Muthusamy Chelliah, Nikesh Garera, Swaprava Nath, Pushpak Bhattacharyya | 2024-04-08T07:15:06Z | cs.CL, cs.AI | In e-commerce, opinion summarization is the process of summarizing the
consensus opinions found in product reviews. However, the potential of
additional sources such as product description and question-answers (QA) has
been considered less often. Moreover, the absence of any supervised training
data makes this task challenging. To address this, we propose a novel synthetic
dataset creation (SDC) strategy that leverages information from reviews as well
as additional sources for selecting one of the reviews as a pseudo-summary to
enable supervised training. Our Multi-Encoder Decoder framework for Opinion
Summarization (MEDOS) employs a separate encoder for each source, enabling
effective selection of information while generating the summary. For
evaluation, due to the unavailability of test sets with additional sources, we
extend the Amazon, Oposum+, and Flipkart test sets and leverage ChatGPT to
annotate summaries. Experiments across nine test sets demonstrate that the
combination of our SDC approach and MEDOS model achieves on average a 14.5%
improvement in ROUGE-1 F1 over the SOTA. Moreover, comparative analysis
underlines the significance of incorporating additional sources for generating
more informative summaries. Human evaluations further indicate that MEDOS
scores relatively higher in coherence and fluency with 0.41 and 0.5 (-1 to 1)
respectively, compared to existing models. To the best of our knowledge, we are
the first to generate opinion summaries leveraging additional sources in a
self-supervised setting. | cs.CL | Product Description and QA Assisted Self-Supervised Opinion Summarization |
2024-04-08T06:40:28Z | http://arxiv.org/pdf/2404.05225v1 | http://arxiv.org/abs/2404.05225v1 | Chuwei Luo, Yufan Shen, Zhaoqing Zhu, Qi Zheng, Zhi Yu, Cong Yao | 2024-04-08T06:40:28Z | cs.CV, cs.CL | Recently, leveraging large language models (LLMs) or multimodal large
language models (MLLMs) for document understanding has been proven very
promising. However, previous works that employ LLMs/MLLMs for document
understanding have not fully explored and utilized the document layout
information, which is vital for precise document understanding. In this paper,
we propose LayoutLLM, an LLM/MLLM based method for document understanding. The
core of LayoutLLM is a layout instruction tuning strategy, which is specially
designed to enhance the comprehension and utilization of document layouts. The
proposed layout instruction tuning strategy consists of two components:
Layout-aware Pre-training and Layout-aware Supervised Fine-tuning. To capture
the characteristics of document layout in Layout-aware Pre-training, three
groups of pre-training tasks, corresponding to document-level, region-level and
segment-level information, are introduced. Furthermore, a novel module called
layout chain-of-thought (LayoutCoT) is devised to enable LayoutLLM to focus on
regions relevant to the question and generate accurate answers. LayoutCoT is
effective for boosting the performance of document understanding. Meanwhile, it
brings a certain degree of interpretability, which could facilitate manual
inspection and correction. Experiments on standard benchmarks show that the
proposed LayoutLLM significantly outperforms existing methods that adopt
open-source 7B LLMs/MLLMs for document understanding. The training data of the
LayoutLLM is publicly available at
https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/DocumentUnderstanding/LayoutLLM | cs.CV | LayoutLLM: Layout Instruction Tuning with Large Language Models for Document Understanding |
2024-04-08T06:35:09Z | http://arxiv.org/pdf/2404.05221v1 | http://arxiv.org/abs/2404.05221v1 | Shibo Hao, Yi Gu, Haotian Luo, Tianyang Liu, Xiyan Shao, Xinyuan Wang, Shuhua Xie, Haodi Ma, Adithya Samavedhi, Qiyue Gao, Zhen Wang, Zhiting Hu | 2024-04-08T06:35:09Z | cs.CL, cs.AI | Generating accurate step-by-step reasoning is essential for Large Language
Models (LLMs) to address complex problems and enhance robustness and
interpretability. Despite the flux of research on developing advanced reasoning
approaches, systematically analyzing the diverse LLMs and reasoning strategies
in generating reasoning chains remains a significant challenge. The
difficulties stem from the lack of two key elements: (1) an automatic method
for evaluating the generated reasoning chains on different tasks, and (2) a
unified formalism and implementation of the diverse reasoning approaches for
systematic comparison. This paper aims to close the gap: (1) We introduce
AutoRace for fully automated reasoning chain evaluation. Existing metrics rely
on expensive human annotations or pre-defined LLM prompts not adaptable to
different tasks. In contrast, AutoRace automatically creates detailed
evaluation criteria tailored for each task, and uses GPT-4 for accurate
evaluation following the criteria. (2) We develop LLM Reasoners, a library for
standardized modular implementation of existing and new reasoning algorithms,
under a unified formulation of the search, reward, and world model components.
With the new evaluation and library, (3) we conduct extensive study of
different reasoning approaches (e.g., CoT, ToT, RAP). The analysis reveals
interesting findings about different factors contributing to reasoning,
including the reward-guidance, breadth-vs-depth in search, world model, and
prompt formats, etc. | cs.CL | LLM Reasoners: New Evaluation, Library, and Analysis of Step-by-Step Reasoning with Large Language Models |
2024-04-08T05:45:03Z | http://arxiv.org/pdf/2404.05210v1 | http://arxiv.org/abs/2404.05210v1 | George Leotescu, Daniel Voinea, Alin-Ionut Popa | 2024-04-08T05:45:03Z | cs.CV, cs.CL, cs.LG | The transformer is a powerful data modelling framework responsible for
remarkable performance on a wide range of tasks. However, they are limited in
terms of scalability as it is suboptimal and inefficient to process
long-sequence data. To this purpose we introduce BLRP (Bidirectional Long-Range
Parser), a novel and versatile attention mechanism designed to increase
performance and efficiency on long-sequence tasks. It leverages short and long
range heuristics in the form of a local sliding window approach combined with a
global bidirectional latent space synthesis technique. We show the benefits and
versatility of our approach on vision and language domains by demonstrating
competitive results against state-of-the-art methods on the Long-Range-Arena
and CIFAR benchmarks together with ablations demonstrating the computational
efficiency. | cs.CV | Bidirectional Long-Range Parser for Sequential Data Understanding |
2024-04-08T04:30:33Z | http://arxiv.org/pdf/2404.05188v1 | http://arxiv.org/abs/2404.05188v1 | Tianshuo Cong, Delong Ran, Zesen Liu, Xinlei He, Jinyuan Liu, Yichen Gong, Qi Li, Anyu Wang, Xiaoyun Wang | 2024-04-08T04:30:33Z | cs.CR, cs.AI, cs.CL | Model merging is a promising lightweight model empowerment technique that
does not rely on expensive computing devices (e.g., GPUs) or require the
collection of specific training data. Instead, it involves editing different
upstream model parameters to absorb their downstream task capabilities.
However, uncertified model merging can infringe upon the Intellectual Property
(IP) rights of the original upstream models. In this paper, we conduct the
first study on the robustness of IP protection methods in model merging
scenarios. We investigate two state-of-the-art IP protection techniques:
Quantization Watermarking and Instructional Fingerprint, along with various
advanced model merging technologies, such as Task Arithmetic, TIES-MERGING, and
so on. Experimental results indicate that current Large Language Model (LLM)
watermarking techniques cannot survive in the merged models, whereas model
fingerprinting techniques can. Our research aims to highlight that model
merging should be an indispensable consideration in the robustness assessment
of model IP protection techniques, thereby promoting the healthy development of
the open-source LLM community. | cs.CR | Have You Merged My Model? On The Robustness of Large Language Model IP Protection Methods Against Model Merging |
2024-04-08T04:14:02Z | http://arxiv.org/pdf/2404.05182v1 | http://arxiv.org/abs/2404.05182v1 | Chao Gao, Sai Qian Zhang | 2024-04-08T04:14:02Z | cs.LG, cs.AI, cs.CL, cs.DC | To enhance the performance of large language models (LLM) on downstream
tasks, one solution is to fine-tune certain LLM parameters and make it better
align with the characteristics of the training dataset. This process is
commonly known as parameter-efficient fine-tuning (PEFT). Due to the scale of
LLM, PEFT operations are usually executed in the public environment (e.g.,
cloud server). This necessitates the sharing of sensitive user data across
public environments, thereby raising potential privacy concerns. To tackle
these challenges, we propose a distributed PEFT framework called DLoRA. DLoRA
enables scalable PEFT operations to be performed collaboratively between the
cloud and user devices. Coupled with the proposed Kill and Revive algorithm,
the evaluation results demonstrate that DLoRA can significantly reduce the
computation and communication workload over the user devices while achieving
superior accuracy and privacy protection. | cs.LG | DLoRA: Distributed Parameter-Efficient Fine-Tuning Solution for Large Language Model |
2024-04-08T03:00:10Z | http://arxiv.org/pdf/2404.05160v1 | http://arxiv.org/abs/2404.05160v1 | Jonathan Crawford | 2024-04-08T03:00:10Z | cs.CL, eess.AS | Parkinsons disease is the second most prevalent neurodegenerative disorder
with over ten million active cases worldwide and one million new diagnoses per
year. Detecting and subsequently diagnosing the disease is challenging because
of symptom heterogeneity with respect to complexity, as well as the type and
timing of phenotypic manifestations. Typically, language impairment can present
in the prodromal phase and precede motor symptoms suggesting that a
linguistic-based approach could serve as a diagnostic method for incipient
Parkinsons disease. Additionally, improved linguistic models may enhance other
approaches through ensemble techniques. The field of large language models is
advancing rapidly, presenting the opportunity to explore the use of these new
models for detecting Parkinsons disease and to improve on current linguistic
approaches with high-dimensional representations of linguistics. We evaluate
the application of state-of-the-art large language models to detect Parkinsons
disease automatically from spontaneous speech with up to 73% accuracy. | cs.CL | Linguistic Changes in Spontaneous Speech for Detecting Parkinsons Disease Using Large Language Models |
2024-04-08T02:55:01Z | http://arxiv.org/pdf/2404.05159v1 | http://arxiv.org/abs/2404.05159v1 | Roopkatha Dey, Aivy Debnath, Sayak Kumar Dutta, Kaustav Ghosh, Arijit Mitra, Arghya Roy Chowdhury, Jaydip Sen | 2024-04-08T02:55:01Z | cs.CL, cs.CR, cs.LG | In various real-world applications such as machine translation, sentiment
analysis, and question answering, a pivotal role is played by NLP models,
facilitating efficient communication and decision-making processes in domains
ranging from healthcare to finance. However, a significant challenge is posed
to the robustness of these natural language processing models by text
adversarial attacks. These attacks involve the deliberate manipulation of input
text to mislead the predictions of the model while maintaining human
interpretability. Despite the remarkable performance achieved by
state-of-the-art models like BERT in various natural language processing tasks,
they are found to remain vulnerable to adversarial perturbations in the input
text. In addressing the vulnerability of text classifiers to adversarial
attacks, three distinct attack mechanisms are explored in this paper using the
victim model BERT: BERT-on-BERT attack, PWWS attack, and Fraud Bargain's Attack
(FBA). Leveraging the IMDB, AG News, and SST2 datasets, a thorough comparative
analysis is conducted to assess the effectiveness of these attacks on the BERT
classifier model. It is revealed by the analysis that PWWS emerges as the most
potent adversary, consistently outperforming other methods across multiple
evaluation scenarios, thereby emphasizing its efficacy in generating
adversarial examples for text classification. Through comprehensive
experimentation, the performance of these attacks is assessed and the findings
indicate that the PWWS attack outperforms others, demonstrating lower runtime,
higher accuracy, and favorable semantic similarity scores. The key insight of
this paper lies in the assessment of the relative performances of three
prevalent state-of-the-art attack mechanisms. | cs.CL | Semantic Stealth: Adversarial Text Attacks on NLP Using Several Methods |
2024-04-08T01:55:28Z | http://arxiv.org/pdf/2404.05144v1 | http://arxiv.org/abs/2404.05144v1 | HyoJe Jung, Yunha Kim, Heejung Choi, Hyeram Seo, Minkyoung Kim, JiYe Han, Gaeun Kee, Seohyun Park, Soyoung Ko, Byeolhee Kim, Suyeon Kim, Tae Joon Jun, Young-Hak Kim | 2024-04-08T01:55:28Z | cs.CL, cs.CV, cs.LG | Medical documentation, including discharge notes, is crucial for ensuring
patient care quality, continuity, and effective medical communication. However,
the manual creation of these documents is not only time-consuming but also
prone to inconsistencies and potential errors. The automation of this
documentation process using artificial intelligence (AI) represents a promising
area of innovation in healthcare. This study directly addresses the
inefficiencies and inaccuracies in creating discharge notes manually,
particularly for cardiac patients, by employing AI techniques, specifically
large language model (LLM). Utilizing a substantial dataset from a cardiology
center, encompassing wide-ranging medical records and physician assessments,
our research evaluates the capability of LLM to enhance the documentation
process. Among the various models assessed, Mistral-7B distinguished itself by
accurately generating discharge notes that significantly improve both
documentation efficiency and the continuity of care for patients. These notes
underwent rigorous qualitative evaluation by medical expert, receiving high
marks for their clinical relevance, completeness, readability, and contribution
to informed decision-making and care planning. Coupled with quantitative
analyses, these results confirm Mistral-7B's efficacy in distilling complex
medical information into concise, coherent summaries. Overall, our findings
illuminate the considerable promise of specialized LLM, such as Mistral-7B, in
refining healthcare documentation workflows and advancing patient care. This
study lays the groundwork for further integrating advanced AI technologies in
healthcare, demonstrating their potential to revolutionize patient
documentation and support better care outcomes. | cs.CL | Enhancing Clinical Efficiency through LLM: Discharge Note Generation for Cardiac Patients |
2024-04-08T01:54:28Z | http://arxiv.org/pdf/2404.05143v1 | http://arxiv.org/abs/2404.05143v1 | Rohan Deepak Ajwani, Zining Zhu, Jonathan Rose, Frank Rudzicz | 2024-04-08T01:54:28Z | cs.CL, cs.AI, cs.LG | Transformer-based Large Language Models (LLMs) have shown exceptional
language generation capabilities in response to text-based prompts. However,
controlling the direction of generation via textual prompts has been
challenging, especially with smaller models. In this work, we explore the use
of Prompt Tuning to achieve controlled language generation. Generated text is
steered using prompt embeddings, which are trained using a small language
model, used as a discriminator. Moreover, we demonstrate that these prompt
embeddings can be trained with a very small dataset, with as low as a few
hundred training examples. Our method thus offers a data and parameter
efficient solution towards controlling language model outputs. We carry out
extensive evaluation on four datasets: SST-5 and Yelp (sentiment analysis),
GYAFC (formality) and JIGSAW (toxic language). Finally, we demonstrate the
efficacy of our method towards mitigating harmful, toxic, and biased text
generated by language models. | cs.CL | Plug and Play with Prompts: A Prompt Tuning Approach for Controlling Text Generation |
2024-04-08T01:21:11Z | http://arxiv.org/pdf/2404.05133v1 | http://arxiv.org/abs/2404.05133v1 | Francesca Grasso, Stefano Locci, Giovanni Siragusa, Luigi Di Caro | 2024-04-08T01:21:11Z | cs.CL | Anthropogenic ecological crisis constitutes a significant challenge that all
within the academy must urgently face, including the Natural Language
Processing (NLP) community. While recent years have seen increasing work
revolving around climate-centric discourse, crucial environmental and
ecological topics outside of climate change remain largely unaddressed, despite
their prominent importance. Mainstream NLP tasks, such as sentiment analysis,
dominate the scene, but there remains an untouched space in the literature
involving the analysis of environmental impacts of certain events and
practices. To address this gap, this paper presents EcoVerse, an annotated
English Twitter dataset of 3,023 tweets spanning a wide spectrum of
environmental topics. We propose a three-level annotation scheme designed for
Eco-Relevance Classification, Stance Detection, and introducing an original
approach for Environmental Impact Analysis. We detail the data collection,
filtering, and labeling process that led to the creation of the dataset.
Remarkable Inter-Annotator Agreement indicates that the annotation scheme
produces consistent annotations of high quality. Subsequent classification
experiments using BERT-based models, including ClimateBERT, are presented.
These yield encouraging results, while also indicating room for a model
specifically tailored for environmental texts. The dataset is made freely
available to stimulate further research. | cs.CL | EcoVerse: An Annotated Twitter Dataset for Eco-Relevance Classification, Environmental Impact Analysis, and Stance Detection |
2024-04-08T00:08:29Z | http://arxiv.org/pdf/2404.08680v1 | http://arxiv.org/abs/2404.08680v1 | Teo Susnjak, Peter Hwang, Napoleon H. Reyes, Andre L. C. Barczak, Timothy R. McIntosh, Surangika Ranathunga | 2024-04-08T00:08:29Z | cs.CL, cs.DL, cs.IR | This research pioneers the use of fine-tuned Large Language Models (LLMs) to
automate Systematic Literature Reviews (SLRs), presenting a significant and
novel contribution in integrating AI to enhance academic research
methodologies. Our study employed the latest fine-tuning methodologies together
with open-sourced LLMs, and demonstrated a practical and efficient approach to
automating the final execution stages of an SLR process that involves knowledge
synthesis. The results maintained high fidelity in factual accuracy in LLM
responses, and were validated through the replication of an existing
PRISMA-conforming SLR. Our research proposed solutions for mitigating LLM
hallucination and proposed mechanisms for tracking LLM responses to their
sources of information, thus demonstrating how this approach can meet the
rigorous demands of scholarly research. The findings ultimately confirmed the
potential of fine-tuned LLMs in streamlining various labor-intensive processes
of conducting literature reviews. Given the potential of this approach and its
applicability across all research domains, this foundational study also
advocated for updating PRISMA reporting guidelines to incorporate AI-driven
processes, ensuring methodological transparency and reliability in future SLRs.
This study broadens the appeal of AI-enhanced tools across various academic and
research fields, setting a new standard for conducting comprehensive and
accurate literature reviews with more efficiency in the face of ever-increasing
volumes of academic studies. | cs.CL | Automating Research Synthesis with Domain-Specific Large Language Model Fine-Tuning |
2024-04-07T22:16:50Z | http://arxiv.org/pdf/2404.05091v4 | http://arxiv.org/abs/2404.05091v4 | Kai Sun, Yushi Bai, Ji Qi, Lei Hou, Juanzi Li | 2024-07-02T12:46:23Z | cs.CL | To advance the evaluation of multimodal math reasoning in large multimodal
models (LMMs), this paper introduces a novel benchmark, MM-MATH. MM-MATH
consists of 5,929 open-ended middle school math problems with visual contexts,
with fine-grained classification across difficulty, grade level, and knowledge
points. Unlike existing benchmarks relying on binary answer comparison, MM-MATH
incorporates both outcome and process evaluations. Process evaluation employs
LMM-as-a-judge to automatically analyze solution steps, identifying and
categorizing errors into specific error types. Extensive evaluation of ten
models on MM-MATH reveals significant challenges for existing LMMs,
highlighting their limited utilization of visual information and struggles with
higher-difficulty problems. The best-performing model achieves only 31%
accuracy on MM-MATH, compared to 82% for humans. This highlights the
challenging nature of our benchmark for existing models and the significant gap
between the multimodal reasoning capabilities of current models and humans. Our
process evaluation reveals that diagram misinterpretation is the most common
error, accounting for more than half of the total error cases, underscoring the
need for improved image comprehension in multimodal reasoning. | cs.CL | MM-MATH: Advancing Multimodal Math Evaluation with Process Evaluation and Fine-grained Classification |
2024-04-07T22:15:13Z | http://arxiv.org/pdf/2404.05090v1 | http://arxiv.org/abs/2404.05090v1 | Mohamed El Amine Seddik, Suei-Wen Chen, Soufiane Hayou, Pierre Youssef, Merouane Debbah | 2024-04-07T22:15:13Z | cs.LG, cs.AI, cs.CL | The phenomenon of model collapse, introduced in (Shumailov et al., 2023),
refers to the deterioration in performance that occurs when new models are
trained on synthetic data generated from previously trained models. This
recursive training loop makes the tails of the original distribution disappear,
thereby making future-generation models forget about the initial (real)
distribution. With the aim of rigorously understanding model collapse in
language models, we consider in this paper a statistical model that allows us
to characterize the impact of various recursive training scenarios.
Specifically, we demonstrate that model collapse cannot be avoided when
training solely on synthetic data. However, when mixing both real and synthetic
data, we provide an estimate of a maximal amount of synthetic data below which
model collapse can eventually be avoided. Our theoretical conclusions are
further supported by empirical validations. | cs.LG | How Bad is Training on Synthetic Data? A Statistical Analysis of Language Model Collapse |
2024-04-07T22:13:43Z | http://arxiv.org/pdf/2404.05089v1 | http://arxiv.org/abs/2404.05089v1 | Alexandre Muzio, Alex Sun, Churan He | 2024-04-07T22:13:43Z | cs.CL, cs.LG | The advancement of deep learning has led to the emergence of
Mixture-of-Experts (MoEs) models, known for their dynamic allocation of
computational resources based on input. Despite their promise, MoEs face
challenges, particularly in terms of memory requirements. To address this, our
work introduces SEER-MoE, a novel two-stage framework for reducing both the
memory footprint and compute requirements of pre-trained MoE models. The first
stage involves pruning the total number of experts using a heavy-hitters
counting guidance, while the second stage employs a regularization-based
fine-tuning strategy to recover accuracy loss and reduce the number of
activated experts during inference. Our empirical studies demonstrate the
effectiveness of our method, resulting in a sparse MoEs model optimized for
inference efficiency with minimal accuracy trade-offs. | cs.CL | SEER-MoE: Sparse Expert Efficiency through Regularization for Mixture-of-Experts |
2024-04-07T22:06:19Z | http://arxiv.org/pdf/2404.05088v1 | http://arxiv.org/abs/2404.05088v1 | Ishani Mondal, Abhilasha Sancheti | 2024-04-07T22:06:19Z | cs.CL | In this paper, we assess the robustness (reliability) of ChatGPT under input
perturbations for one of the most fundamental tasks of Information Extraction
(IE) i.e. Named Entity Recognition (NER). Despite the hype, the majority of the
researchers have vouched for its language understanding and generation
capabilities; a little attention has been paid to understand its robustness:
How the input-perturbations affect 1) the predictions, 2) the confidence of
predictions and 3) the quality of rationale behind its prediction. We perform a
systematic analysis of ChatGPT's robustness (under both zero-shot and few-shot
setup) on two NER datasets using both automatic and human evaluation. Based on
automatic evaluation metrics, we find that 1) ChatGPT is more brittle on Drug
or Disease replacements (rare entities) compared to the perturbations on widely
known Person or Location entities, 2) the quality of explanations for the same
entity considerably differ under different types of "Entity-Specific" and
"Context-Specific" perturbations and the quality can be significantly improved
using in-context learning, and 3) it is overconfident for majority of the
incorrect predictions, and hence it could lead to misguidance of the end-users. | cs.CL | How much reliable is ChatGPT's prediction on Information Extraction under Input Perturbations? |
2024-04-07T22:00:50Z | http://arxiv.org/pdf/2404.05086v1 | http://arxiv.org/abs/2404.05086v1 | Vlad Fomenko, Han Yu, Jongho Lee, Stanley Hsieh, Weizhu Chen | 2024-04-07T22:00:50Z | cs.LG, cs.AI, cs.CL | LoRA (Low-Rank Adaptation) has emerged as a preferred method for efficiently
adapting Large Language Models (LLMs) with remarkable simplicity and efficacy.
This note extends the original LoRA paper by offering new perspectives that
were not initially discussed and presents a series of insights for deploying
LoRA at scale. Without introducing new experiments, we aim to improve the
understanding and application of LoRA. | cs.LG | A Note on LoRA |
2024-04-07T21:46:47Z | http://arxiv.org/pdf/2404.05083v1 | http://arxiv.org/abs/2404.05083v1 | Yimu Wang, Shuai Yuan, Xiangru Jian, Wei Pang, Mushi Wang, Ning Yu | 2024-04-07T21:46:47Z | cs.CV, cs.CL, cs.IR, cs.LG | While recent progress in video-text retrieval has been driven by the
exploration of powerful model architectures and training strategies, the
representation learning ability of video-text retrieval models is still limited
due to low-quality and scarce training data annotations. To address this issue,
we present a novel video-text learning paradigm, HaVTR, which augments video
and text data to learn more generalized features. Specifically, we first adopt
a simple augmentation method, which generates self-similar data by randomly
duplicating or dropping subwords and frames. In addition, inspired by the
recent advancement in visual and language generative models, we propose a more
powerful augmentation method through textual paraphrasing and video stylization
using large language models (LLMs) and visual generative models (VGMs).
Further, to bring richer information into video and text, we propose a
hallucination-based augmentation method, where we use LLMs and VGMs to generate
and add new relevant information to the original data. Benefiting from the
enriched data, extensive experiments on several video-text retrieval benchmarks
demonstrate the superiority of HaVTR over existing methods. | cs.CV | HaVTR: Improving Video-Text Retrieval Through Augmentation Using Large Foundation Models |
2024-04-07T21:02:55Z | http://arxiv.org/pdf/2404.05073v1 | http://arxiv.org/abs/2404.05073v1 | Stefano Scanzio, Gianluca Cena, Adriano Valenzano | 2024-04-07T21:02:55Z | cs.NI, cs.CL | Embedding a programming language in a QR code is a new and extremely
promising opportunity, as it makes devices and objects smarter without
necessarily requiring an Internet connection. In this paper, all the steps
needed to translate a program written in a high-level programming language to
its binary representation encoded in a QR code, and the opposite process that,
starting from the QR code, executes it by means of a virtual machine, have been
carefully detailed. The proposed programming language was named QRscript, and
can be easily extended so as to integrate new features. One of the main design
goals was to produce a very compact target binary code. In particular, in this
work we propose a specific sub-language (a dialect) that is aimed at encoding
decision trees. Besides industrial scenarios, this is useful in many other
application fields. The reported example, related to the configuration of an
industrial networked device, highlights the potential of the proposed
technology, and permits to better understand all the translation steps. | cs.NI | QRscript: Embedding a Programming Language in QR codes to support Decision and Management |
2024-04-07T19:00:45Z | http://arxiv.org/pdf/2404.05046v1 | http://arxiv.org/abs/2404.05046v1 | Liqiang Jing, Xinya Du | 2024-04-07T19:00:45Z | cs.CV, cs.CL | Large Vision-Language Models (LVLMs) have demonstrated proficiency in
tackling a variety of visual-language tasks. However, current LVLMs suffer from
misalignment between text and image modalities which causes three kinds of
hallucination problems, i.e., object existence, object attribute, and object
relationship. To tackle this issue, existing methods mainly utilize
Reinforcement Learning (RL) to align modalities in LVLMs. However, they still
suffer from three main limitations: (1) General feedback can not indicate the
hallucination type contained in the response; (2) Sparse rewards only give the
sequence-level reward for the whole response; and (3)Annotation cost is
time-consuming and labor-intensive. To handle these limitations, we propose an
innovative method to align modalities in LVLMs through Fine-Grained Artificial
Intelligence Feedback (FGAIF), which mainly consists of three steps: AI-based
Feedback Collection, Fine-grained Reward Model Training, and Reinforcement
Learning with Fine-grained Reward. Specifically, We first utilize AI tools to
predict the types of hallucination for each segment in the response and obtain
a collection of fine-grained feedback. Then, based on the collected reward
data, three specialized reward models are trained to produce dense rewards.
Finally, a novel fine-grained feedback module is integrated into the Proximal
Policy Optimization (PPO) algorithm. Extensive experiments are conducted on
hallucination and general benchmarks, demonstrating the superior performance of
our proposed method. Notably, compared with previous models trained with the
RL-based aligning method, our proposed method is effective even with fewer
parameters. | cs.CV | FGAIF: Aligning Large Vision-Language Models with Fine-grained AI Feedback |
2024-04-07T17:17:23Z | http://arxiv.org/pdf/2404.05019v1 | http://arxiv.org/abs/2404.05019v1 | Weilin Cai, Juyong Jiang, Le Qin, Junwei Cui, Sunghun Kim, Jiayi Huang | 2024-04-07T17:17:23Z | cs.LG, cs.CL, cs.DC | Expert parallelism has been introduced as a strategy to distribute the
computational workload of sparsely-gated mixture-of-experts (MoE) models across
multiple computing devices, facilitating the execution of these increasingly
large-scale models. However, the All-to-All communication intrinsic to expert
parallelism constitutes a significant overhead, diminishing the MoE models'
efficiency. Current optimization approaches offer some relief, yet they are
constrained by the sequential interdependence of communication and computation
operations. To address this limitation, we present a novel shortcut-connected
MoE architecture with overlapping parallel strategy, designated as ScMoE, which
effectively decouples communication from its conventional sequence, allowing
for a substantial overlap of 70% to 100% with computation. When compared with
the prevalent top-2 MoE architecture, ScMoE demonstrates training speed
improvements of 30% and 11%, and inference improvements of 40% and 15%, in our
PCIe and NVLink hardware environments, respectively, where communication
constitutes 60% and 15% of the total MoE time consumption. On the other hand,
extensive experiments and theoretical analyses indicate that ScMoE not only
achieves comparable but in some instances surpasses the model quality of
existing approaches in vision and language tasks. | cs.LG | Shortcut-connected Expert Parallelism for Accelerating Mixture-of-Experts |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.