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
|
|---|---|---|---|---|---|---|---|---|
2023-11-17T20:48:58Z | http://arxiv.org/pdf/2311.10862v1 | http://arxiv.org/abs/2311.10862v1 | Sergei O. Kuznetsov, Vasilii A. Gromov, Nikita S. Borodin, Andrei M. Divavin | 2023-11-17T20:48:58Z | cs.CL, cs.AI, math.AT | Some results of a computational experiment for determining the intrinsic
dimension of linguistic varieties for the Bengali and Russian languages are
presented. At the same time, both sets of words and sets of bigrams in these
languages were considered separately. The method used to solve this problem was
based on formal concept analysis algorithms. It was found that the intrinsic
dimensions of these languages are significantly less than the dimensions used
in popular neural network models in natural language processing. | cs.CL | Formal concept analysis for evaluating intrinsic dimension of a natural language |
2023-11-17T20:07:54Z | http://arxiv.org/pdf/2311.10847v2 | http://arxiv.org/abs/2311.10847v2 | Joshua Belofsky | 2023-12-01T06:38:18Z | cs.CL, cs.LG | This paper introduces a method for adapting LoRA adapters in smaller-sized
language models to arbitrary downstream tasks. Unlike standard
mixture-of-expert architectures, our method employs a gradient-free routing
function to choose a weighted combination of experts without increasing the
compute requirements for training or inference. The results show that
token-level adaptation of LoRA adapters outperforms the base Llama-2-7b model
across mathematical (GSM8K), scientific (ARC-Challenge), reading comprehension
(SQuAD), and coding (CodeAlpaca-20k) tasks. Further evaluations also show that
the average performance of token-level adaptation outperforms individual models
fine-tuned for each of the tasks with the best performance observed in
adaptation of every-other token during inference. The code for this study is
made available through a public repository. | cs.CL | Token-Level Adaptation of LoRA Adapters for Downstream Task Generalization |
2023-11-17T18:59:56Z | http://arxiv.org/pdf/2311.10813v3 | http://arxiv.org/abs/2311.10813v3 | Jiageng Mao, Junjie Ye, Yuxi Qian, Marco Pavone, Yue Wang | 2023-11-27T20:53:35Z | cs.CV, cs.AI, cs.CL, cs.RO | Human-level driving is an ultimate goal of autonomous driving. Conventional
approaches formulate autonomous driving as a perception-prediction-planning
framework, yet their systems do not capitalize on the inherent reasoning
ability and experiential knowledge of humans. In this paper, we propose a
fundamental paradigm shift from current pipelines, exploiting Large Language
Models (LLMs) as a cognitive agent to integrate human-like intelligence into
autonomous driving systems. Our approach, termed Agent-Driver, transforms the
traditional autonomous driving pipeline by introducing a versatile tool library
accessible via function calls, a cognitive memory of common sense and
experiential knowledge for decision-making, and a reasoning engine capable of
chain-of-thought reasoning, task planning, motion planning, and
self-reflection. Powered by LLMs, our Agent-Driver is endowed with intuitive
common sense and robust reasoning capabilities, thus enabling a more nuanced,
human-like approach to autonomous driving. We evaluate our approach on the
large-scale nuScenes benchmark, and extensive experiments substantiate that our
Agent-Driver significantly outperforms the state-of-the-art driving methods by
a large margin. Our approach also demonstrates superior interpretability and
few-shot learning ability to these methods. Code will be released. | cs.CV | A Language Agent for Autonomous Driving |
2023-11-17T18:45:45Z | http://arxiv.org/pdf/2311.10702v2 | http://arxiv.org/abs/2311.10702v2 | Hamish Ivison, Yizhong Wang, Valentina Pyatkin, Nathan Lambert, Matthew Peters, Pradeep Dasigi, Joel Jang, David Wadden, Noah A. Smith, Iz Beltagy, Hannaneh Hajishirzi | 2023-11-20T02:01:33Z | cs.CL | Since the release of T\"ULU [Wang et al., 2023b], open resources for
instruction tuning have developed quickly, from better base models to new
finetuning techniques. We test and incorporate a number of these advances into
T\"ULU, resulting in T\"ULU 2, a suite of improved T\"ULU models for advancing
the understanding and best practices of adapting pretrained language models to
downstream tasks and user preferences. Concretely, we release: (1)
T\"ULU-V2-mix, an improved collection of high-quality instruction datasets; (2)
T\"ULU 2, LLAMA-2 models finetuned on the V2 mixture; (3) T\"ULU 2+DPO, T\"ULU
2 models trained with direct preference optimization (DPO), including the
largest DPO-trained model to date (T\"ULU 2+DPO 70B); (4) CODE T\"ULU 2, CODE
LLAMA models finetuned on our V2 mix that outperform CODE LLAMA and its
instruction-tuned variant, CODE LLAMA-Instruct. Our evaluation from multiple
perspectives shows that the T\"ULU 2 suite achieves state-of-the-art
performance among open models and matches or exceeds the performance of
GPT-3.5-turbo-0301 on several benchmarks. We release all the checkpoints, data,
training and evaluation code to facilitate future open efforts on adapting
large language models. | cs.CL | Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2 |
2023-11-17T18:32:17Z | http://arxiv.org/pdf/2311.10697v1 | http://arxiv.org/abs/2311.10697v1 | Keivalya Pandya | 2023-11-17T18:32:17Z | cs.CL, cs.AI, cs.IR, cs.LG | Chat models are capable of answering a wide range of questions, however, the
accuracy of their responses is highly uncertain. In this research, we propose a
specialized PEFT-MedAware model where we utilize parameter-efficient
fine-tuning (PEFT) to enhance the Falcon-1b large language model on specialized
MedQuAD data consisting of 16,407 medical QA pairs, leveraging only 0.44% of
its trainable parameters to enhance computational efficiency. The paper adopts
data preprocessing and PEFT to optimize model performance, complemented by a
BitsAndBytesConfig for efficient transformer training. The resulting model was
capable of outperforming other LLMs in medical question-answering tasks in
specific domains with greater accuracy utilizing limited computational
resources making it suitable for deployment in resource-constrained
environments. We propose further improvements through expanded datasets, larger
models, and feedback mechanisms for sustained medical relevancy. Our work
highlights the efficiency gains and specialized capabilities of PEFT in medical
AI, outpacing standard models in precision without extensive resource demands.
The proposed model and data are released for research purposes only. | cs.CL | PEFT-MedAware: Large Language Model for Medical Awareness |
2023-11-17T18:14:08Z | http://arxiv.org/pdf/2311.10810v1 | http://arxiv.org/abs/2311.10810v1 | Yao-Shun Chuang, Xiaoqian Jiang, Chun-Teh Lee, Ryan Brandon, Duong Tran, Oluwabunmi Tokede, Muhammad F. Walji | 2023-11-17T18:14:08Z | cs.CL, cs.AI | This study explored the usability of prompt generation on named entity
recognition (NER) tasks and the performance in different settings of the
prompt. The prompt generation by GPT-J models was utilized to directly test the
gold standard as well as to generate the seed and further fed to the RoBERTa
model with the spaCy package. In the direct test, a lower ratio of negative
examples with higher numbers of examples in prompt achieved the best results
with a F1 score of 0.72. The performance revealed consistency, 0.92-0.97 in the
F1 score, in all settings after training with the RoBERTa model. The study
highlighted the importance of seed quality rather than quantity in feeding NER
models. This research reports on an efficient and accurate way to mine clinical
notes for periodontal diagnoses, allowing researchers to easily and quickly
build a NER model with the prompt generation approach. | cs.CL | Use GPT-J Prompt Generation with RoBERTa for NER Models on Diagnosis Extraction of Periodontal Diagnosis from Electronic Dental Records |
2023-11-17T16:58:52Z | http://arxiv.org/pdf/2311.10642v4 | http://arxiv.org/abs/2311.10642v4 | Vukasin Bozic, Danilo Dordevic, Daniele Coppola, Joseph Thommes, Sidak Pal Singh | 2024-02-04T20:39:33Z | cs.CL, cs.LG | This work presents an analysis of the effectiveness of using standard shallow
feed-forward networks to mimic the behavior of the attention mechanism in the
original Transformer model, a state-of-the-art architecture for
sequence-to-sequence tasks. We substitute key elements of the attention
mechanism in the Transformer with simple feed-forward networks, trained using
the original components via knowledge distillation. Our experiments, conducted
on the IWSLT2017 dataset, reveal the capacity of these "attentionless
Transformers" to rival the performance of the original architecture. Through
rigorous ablation studies, and experimenting with various replacement network
types and sizes, we offer insights that support the viability of our approach.
This not only sheds light on the adaptability of shallow feed-forward networks
in emulating attention mechanisms but also underscores their potential to
streamline complex architectures for sequence-to-sequence tasks. | cs.CL | Rethinking Attention: Exploring Shallow Feed-Forward Neural Networks as an Alternative to Attention Layers in Transformers |
2023-11-17T16:09:10Z | http://arxiv.org/pdf/2311.10614v1 | http://arxiv.org/abs/2311.10614v1 | Ruohong Zhang, Luyu Gao, Chen Zheng, Zhen Fan, Guokun Lai, Zheng Zhang, Fangzhou Ai, Yiming Yang, Hongxia Yang | 2023-11-17T16:09:10Z | cs.CL, cs.AI | Large Language Models (LLMs), despite their great power in language
generation, often encounter challenges when dealing with intricate and
knowledge-demanding queries in specific domains. This paper introduces a novel
approach to enhance LLMs by effectively extracting the relevant knowledge from
domain-specific textual sources, and the adaptive training of a chatbot with
domain-specific inquiries. Our two-step approach starts from training a
knowledge miner, namely LLMiner, which autonomously extracts Question-Answer
pairs from relevant documents through a chain-of-thought reasoning process.
Subsequently, we blend the mined QA pairs with a conversational dataset to
fine-tune the LLM as a chatbot, thereby enriching its domain-specific expertise
and conversational capabilities. We also developed a new evaluation benchmark
which comprises four domain-specific text corpora and associated human-crafted
QA pairs for testing. Our model shows remarkable performance improvement over
generally aligned LLM and surpasses domain-adapted models directly fine-tuned
on domain corpus. In particular, LLMiner achieves this with minimal human
intervention, requiring only 600 seed instances, thereby providing a pathway
towards self-improvement of LLMs through model-synthesized training data. | cs.CL | A Self-enhancement Approach for Domain-specific Chatbot Training via Knowledge Mining and Digest |
2023-11-17T15:49:11Z | http://arxiv.org/pdf/2311.10596v1 | http://arxiv.org/abs/2311.10596v1 | Steven Leung, Filippos Papapolyzos | 2023-11-17T15:49:11Z | cs.CL, cs.AI | Personal attacks in the context of social media conversations often lead to
fast-paced derailment, leading to even more harmful exchanges being made.
State-of-the-art systems for the detection of such conversational derailment
often make use of deep learning approaches for prediction purposes. In this
paper, we show that an Attention-based BERT architecture, pre-trained on a
large Twitter corpus and fine-tuned on our task, is efficient and effective in
making such predictions. This model shows clear advantages in performance to
the existing LSTM model we use as a baseline. Additionally, we show that this
impressive performance can be attained through fine-tuning on a relatively
small, novel dataset, particularly after mitigating overfitting issues through
synthetic oversampling techniques. By introducing the first transformer based
model for forecasting conversational events on Twitter, this work lays the
foundation for a practical tool to encourage better interactions on one of the
most ubiquitous social media platforms. | cs.CL | Hashing it Out: Predicting Unhealthy Conversations on Twitter |
2023-11-17T15:37:18Z | http://arxiv.org/pdf/2311.10587v1 | http://arxiv.org/abs/2311.10587v1 | Daniel Russo, Shane Peter Kaszefski-Yaschuk, Jacopo Staiano, Marco Guerini | 2023-11-17T15:37:18Z | cs.CL | The proliferation of misinformation on social media platforms (SMPs) poses a
significant danger to public health, social cohesion and ultimately democracy.
Previous research has shown how social correction can be an effective way to
curb misinformation, by engaging directly in a constructive dialogue with users
who spread -- often in good faith -- misleading messages. Although professional
fact-checkers are crucial to debunking viral claims, they usually do not engage
in conversations on social media. Thereby, significant effort has been made to
automate the use of fact-checker material in social correction; however, no
previous work has tried to integrate it with the style and pragmatics that are
commonly employed in social media communication. To fill this gap, we present
VerMouth, the first large-scale dataset comprising roughly 12 thousand
claim-response pairs (linked to debunking articles), accounting for both
SMP-style and basic emotions, two factors which have a significant role in
misinformation credibility and spreading. To collect this dataset we used a
technique based on an author-reviewer pipeline, which efficiently combines LLMs
and human annotators to obtain high-quality data. We also provide comprehensive
experiments showing how models trained on our proposed dataset have significant
improvements in terms of output quality and generalization capabilities. | cs.CL | Countering Misinformation via Emotional Response Generation |
2023-11-17T14:08:44Z | http://arxiv.org/pdf/2311.10541v1 | http://arxiv.org/abs/2311.10541v1 | Fatima Muhammad Adam, Abubakar Yakubu Zandam, Isa Inuwa-Dutse | 2023-11-17T14:08:44Z | cs.CL | Hausa is a major Chadic language, spoken by over 100 million people in
Africa. However, from a computational linguistic perspective, it is considered
a low-resource language, with limited resources to support Natural Language
Processing (NLP) tasks. Online platforms often facilitate social interactions
that can lead to the use of offensive and threatening language, which can go
undetected due to the lack of detection systems designed for Hausa. This study
aimed to address this issue by (1) conducting two user studies (n=308) to
investigate cyberbullying-related issues, (2) collecting and annotating the
first set of offensive and threatening datasets to support relevant downstream
tasks in Hausa, (3) developing a detection system to flag offensive and
threatening content, and (4) evaluating the detection system and the efficacy
of the Google-based translation engine in detecting offensive and threatening
terms in Hausa. We found that offensive and threatening content is quite
common, particularly when discussing religion and politics. Our detection
system was able to detect more than 70% of offensive and threatening content,
although many of these were mistranslated by Google's translation engine. We
attribute this to the subtle relationship between offensive and threatening
content and idiomatic expressions in the Hausa language. We recommend that
diverse stakeholders participate in understanding local conventions and
demographics in order to develop a more effective detection system. These
insights are essential for implementing targeted moderation strategies to
create a safe and inclusive online environment. | cs.CL | Detection of Offensive and Threatening Online Content in a Low Resource Language |
2023-11-17T13:35:10Z | http://arxiv.org/pdf/2311.10514v1 | http://arxiv.org/abs/2311.10514v1 | Stetsenko Daria | 2023-11-17T13:35:10Z | cs.CL | Many under-resourced languages require high-quality datasets for specific
tasks such as offensive language detection, disinformation, or misinformation
identification. However, the intricacies of the content may have a detrimental
effect on the annotators. The article aims to revisit an approach of
pseudo-labeling sensitive data on the example of Ukrainian tweets covering the
Russian-Ukrainian war. Nowadays, this acute topic is in the spotlight of
various language manipulations that cause numerous disinformation and profanity
on social media platforms. The conducted experiment highlights three main
stages of data annotation and underlines the main obstacles during machine
annotation. Ultimately, we provide a fundamental statistical analysis of the
obtained data, evaluation of models used for pseudo-labelling, and set further
guidelines on how the scientists can leverage the corpus to execute more
advanced research and extend the existing data samples without annotators'
engagement. | cs.CL | When a Language Question Is at Stake. A Revisited Approach to Label Sensitive Content |
2023-11-17T13:10:58Z | http://arxiv.org/pdf/2311.10505v1 | http://arxiv.org/abs/2311.10505v1 | Simone Caruso, Carmine Dodaro, Marco Maratea, Marco Mochi, Francesco Riccio | 2023-11-17T13:10:58Z | cs.AI, cs.CL, cs.LO | Answer Set Programming (ASP) is a popular declarative programming language
for solving hard combinatorial problems. Although ASP has gained widespread
acceptance in academic and industrial contexts, there are certain user groups
who may find it more advantageous to employ a higher-level language that
closely resembles natural language when specifying ASP programs. In this paper,
we propose a novel tool, called CNL2ASP, for translating English sentences
expressed in a controlled natural language (CNL) form into ASP. In particular,
we first provide a definition of the type of sentences allowed by our CNL and
their translation as ASP rules, and then exemplify the usage of the CNL for the
specification of both synthetic and real-world combinatorial problems. Finally,
we report the results of an experimental analysis conducted on the real-world
problems to compare the performance of automatically generated encodings with
the ones written by ASP practitioners, showing that our tool can obtain
satisfactory performance on these benchmarks. Under consideration in Theory and
Practice of Logic Programming (TPLP). | cs.AI | CNL2ASP: converting controlled natural language sentences into ASP |
2023-11-17T13:07:00Z | http://arxiv.org/pdf/2311.10804v1 | http://arxiv.org/abs/2311.10804v1 | Mathias Vogel | 2023-11-17T13:07:00Z | cs.CL, cs.AI | This report explores the challenge of enhancing expressiveness control in
Text-to-Speech (TTS) models by augmenting a frozen pretrained model with a
Diffusion Model that is conditioned on joint semantic audio/text embeddings.
The paper identifies the challenges encountered when working with a VAE-based
TTS model and evaluates different image-to-image methods for altering latent
speech features. Our results offer valuable insights into the complexities of
adding expressiveness control to TTS systems and open avenues for future
research in this direction. | cs.CL | A Study on Altering the Latent Space of Pretrained Text to Speech Models for Improved Expressiveness |
2023-11-17T10:14:45Z | http://arxiv.org/pdf/2311.10436v1 | http://arxiv.org/abs/2311.10436v1 | Kasun Wickramasinghe, Nisansa de Silva | 2023-11-17T10:14:45Z | cs.CL | Since their inception, embeddings have become a primary ingredient in many
flavours of Natural Language Processing (NLP) tasks supplanting earlier types
of representation. Even though multilingual embeddings have been used for the
increasing number of multilingual tasks, due to the scarcity of parallel
training data, low-resource languages such as Sinhala, tend to focus more on
monolingual embeddings. Then when it comes to the aforementioned multi-lingual
tasks, it is challenging to utilize these monolingual embeddings given that
even if the embedding spaces have a similar geometric arrangement due to an
identical training process, the embeddings of the languages considered are not
aligned. This is solved by the embedding alignment task. Even in this,
high-resource language pairs are in the limelight while low-resource languages
such as Sinhala which is in dire need of help seem to have fallen by the
wayside. In this paper, we try to align Sinhala and English word embedding
spaces based on available alignment techniques and introduce a benchmark for
Sinhala language embedding alignment. In addition to that, to facilitate the
supervised alignment, as an intermediate task, we also introduce
Sinhala-English alignment datasets. These datasets serve as our anchor datasets
for supervised word embedding alignment. Even though we do not obtain results
comparable to the high-resource languages such as French, German, or Chinese,
we believe our work lays the groundwork for more specialized alignment between
English and Sinhala embeddings. | cs.CL | Sinhala-English Word Embedding Alignment: Introducing Datasets and Benchmark for a Low Resource Language |
2023-11-17T10:09:12Z | http://arxiv.org/pdf/2311.10431v1 | http://arxiv.org/abs/2311.10431v1 | Zhengqi He, Taro Toyoizumi | 2023-11-17T10:09:12Z | cs.CL | Understanding how humans process natural language has long been a vital
research direction. The field of natural language processing (NLP) has recently
experienced a surge in the development of powerful language models. These
models have proven to be invaluable tools for studying another complex system
known to process human language: the brain. Previous studies have demonstrated
that the features of language models can be mapped to fMRI brain activity. This
raises the question: is there a commonality between information processing in
language models and the human brain? To estimate information flow patterns in a
language model, we examined the causal relationships between different layers.
Drawing inspiration from the workspace framework for consciousness, we
hypothesized that features integrating more information would more accurately
predict higher hierarchical brain activity. To validate this hypothesis, we
classified language model features into two categories based on causal network
measures: 'low in-degree' and 'high in-degree'. We subsequently compared the
brain prediction accuracy maps for these two groups. Our results reveal that
the difference in prediction accuracy follows a hierarchical pattern,
consistent with the cortical hierarchy map revealed by activity time constants.
This finding suggests a parallel between how language models and the human
brain process linguistic information. | cs.CL | Causal Graph in Language Model Rediscovers Cortical Hierarchy in Human Narrative Processing |
2023-11-17T08:56:13Z | http://arxiv.org/pdf/2311.10395v2 | http://arxiv.org/abs/2311.10395v2 | Yi Yang, Hanyu Duan, Ahmed Abbasi, John P. Lalor, Kar Yan Tam | 2024-06-16T03:33:03Z | cs.CL | Transformer-based pretrained large language models (PLM) such as BERT and GPT
have achieved remarkable success in NLP tasks. However, PLMs are prone to
encoding stereotypical biases. Although a burgeoning literature has emerged on
stereotypical bias mitigation in PLMs, such as work on debiasing gender and
racial stereotyping, how such biases manifest and behave internally within PLMs
remains largely unknown. Understanding the internal stereotyping mechanisms may
allow better assessment of model fairness and guide the development of
effective mitigation strategies. In this work, we focus on attention heads, a
major component of the Transformer architecture, and propose a bias analysis
framework to explore and identify a small set of biased heads that are found to
contribute to a PLM's stereotypical bias. We conduct extensive experiments to
validate the existence of these biased heads and to better understand how they
behave. We investigate gender and racial bias in the English language in two
types of Transformer-based PLMs: the encoder-based BERT model and the
decoder-based autoregressive GPT model. Overall, the results shed light on
understanding the bias behavior in pretrained language models. | cs.CL | Bias A-head? Analyzing Bias in Transformer-Based Language Model Attention Heads |
2023-11-17T07:56:01Z | http://arxiv.org/pdf/2311.10373v1 | http://arxiv.org/abs/2311.10373v1 | Ting Xu, Zhen Wu, Huiyun Yang, Xinyu Dai | 2023-11-17T07:56:01Z | cs.CL | Aspect Sentiment Triplet Extraction (ASTE) has achieved promising results
while relying on sufficient annotation data in a specific domain. However, it
is infeasible to annotate data for each individual domain. We propose to
explore ASTE in the cross-domain setting, which transfers knowledge from a
resource-rich source domain to a resource-poor target domain, thereby
alleviating the reliance on labeled data in the target domain. To effectively
transfer the knowledge across domains and extract the sentiment triplets
accurately, we propose a method named Fine-grained cOntrAstive Learning (FOAL)
to reduce the domain discrepancy and preserve the discriminability of each
category. Experiments on six transfer pairs show that FOAL achieves 6%
performance gains and reduces the domain discrepancy significantly compared
with strong baselines. Our code will be publicly available once accepted. | cs.CL | FOAL: Fine-grained Contrastive Learning for Cross-domain Aspect Sentiment Triplet Extraction |
2023-11-17T07:40:46Z | http://arxiv.org/pdf/2311.10367v1 | http://arxiv.org/abs/2311.10367v1 | Hanyu Duan, Yixuan Tang, Yi Yang, Ahmed Abbasi, Kar Yan Tam | 2023-11-17T07:40:46Z | cs.CL | In-Context Learning (ICL) and Instruction Tuning (IT) are two primary
paradigms of adopting Large Language Models (LLMs) to downstream applications.
However, they are significantly different. In ICL, a set of demonstrations are
provided at inference time but the LLM's parameters are not updated. In IT, a
set of demonstrations are used to tune LLM's parameters in training time but no
demonstrations are used at inference time. Although a growing body of
literature has explored ICL and IT, studies on these topics have largely been
conducted in isolation, leading to a disconnect between these two paradigms. In
this work, we explore the relationship between ICL and IT by examining how the
hidden states of LLMs change in these two paradigms. Through carefully designed
experiments conducted with LLaMA-2 (7B and 13B), we find that ICL is implicit
IT. In other words, ICL changes an LLM's hidden states as if the demonstrations
were used to instructionally tune the model. Furthermore, the convergence
between ICL and IT is largely contingent upon several factors related to the
provided demonstrations. Overall, this work offers a unique perspective to
explore the connection between ICL and IT and sheds light on understanding the
behaviors of LLM. | cs.CL | Exploring the Relationship between In-Context Learning and Instruction Tuning |
2023-11-17T06:55:32Z | http://arxiv.org/pdf/2311.10797v2 | http://arxiv.org/abs/2311.10797v2 | Bibek Upadhayay, Vahid Behzadan | 2024-04-05T06:13:15Z | cs.CL, cs.AI | Creating multilingual LLMs poses a significant challenge. Pretraining or
fine-tuning LLMs to adopt new languages is evidently very costly. Furthermore,
there exist limitations concerning benchmark datasets and the metrics used to
measure model performance in multilingual settings. This paper proposes
cost-effective solutions to both aforementioned challenges. Firstly, we
introduce the Multilingual Instruction-Tuning Dataset (MITS), comprised of
Alpaca-52K, Dolly-15K, and Vicuna Benchmark translations into 132 languages.
Secondly, we propose a new method called \emph{TaCo: Translation-Assisted
Cross-Linguality}, which utilizes translations in a chain-of-thought process to
instruction-tune LLMs on new languages through a curriculum-learning process.
As a proof of concept, we experimented with the instruction-tuned Guanaco-33B
model, performing further instruction tuning using our proposed TaCo method in
three low-resource languages and one high-resource language. Our results
indicate that the TaCo method impresses GPT-4 with an 82\% score for a
low-resource language in the Vicuna Benchmark dataset, doubling the performance
in contrast to instruction tuning alone. Furthermore, TaCo shows promise in
creating multilingual LLMs, even for low-resource languages. We have released
our datasets and model adapters\footnote{https://github.com/UNHSAILLab/TaCo} ,
encouraging the research community to utilize these resources to advance work
on multilingual LLMs. | cs.CL | TaCo: Enhancing Cross-Lingual Transfer for Low-Resource Languages in LLMs through Translation-Assisted Chain-of-Thought Processes |
2023-11-17T06:13:02Z | http://arxiv.org/pdf/2311.10344v1 | http://arxiv.org/abs/2311.10344v1 | Tongquan Zhou, Yao Zhang, Siyi Cao, Yulu Li, Tao Wang | 2023-11-17T06:13:02Z | cs.CL | ChatGPT has shown its great power in text processing, including its reasoning
ability from text reading. However, there has not been any direct comparison
between human readers and ChatGPT in reasoning ability related to text reading.
This study was undertaken to investigate how ChatGPTs (i.e., ChatGPT and
ChatGPT Plus) and Chinese senior school students as ESL learners exhibited
their reasoning ability from English narrative texts. Additionally, we compared
the two ChatGPTs in the reasoning performances when commands were updated
elaborately. The whole study was composed of three reasoning tests: Test 1 for
commonsense inference, Test 2 for emotional inference, and Test 3 for causal
inference. The results showed that in Test 1, the students outdid the two
ChatGPT versions in local-culture-related inferences but performed worse than
the chatbots in daily-life inferences. In Test 2, ChatGPT Plus excelled whereas
ChatGPT lagged behind in accuracy. In association with both accuracy and
frequency of correct responses, the students were inferior to the two chatbots.
Compared with ChatGPTs' better performance in positive emotions, the students
showed their superiority in inferring negative emotions. In Test 3, the
students demonstrated better logical analysis, outdoing both chatbots. In
updating command condition, ChatGPT Plus displayed good causal reasoning
ability while ChatGPT kept unchanged. Our study reveals that human readers and
ChatGPTs have their respective advantages and disadvantages in drawing
inferences from text reading comprehension, unlocking a complementary
relationship in text-based reasoning. | cs.CL | Complementary Advantages of ChatGPTs and Human Readers in Reasoning: Evidence from English Text Reading Comprehension |
2023-11-17T01:33:05Z | http://arxiv.org/pdf/2311.10271v1 | http://arxiv.org/abs/2311.10271v1 | Hong Liu, Yucheng Cai, Yuan Zhou, Zhijian Ou, Yi Huang, Junlan Feng | 2023-11-17T01:33:05Z | cs.CL | Continual learning is crucial for dialog state tracking (DST) in dialog
systems, since requirements from users for new functionalities are often
encountered. However, most of existing continual learning methods for DST
require task identities during testing, which is a severe limit in real-world
applications. In this paper, we aim to address continual learning of DST in the
class-incremental scenario (namely the task identity is unknown in testing).
Inspired by the recently emerging prompt tuning method that performs well on
dialog systems, we propose to use the prompt pool method, where we maintain a
pool of key-value paired prompts and select prompts from the pool according to
the distance between the dialog history and the prompt keys. The proposed
method can automatically identify tasks and select appropriate prompts during
testing. We conduct experiments on Schema-Guided Dialog dataset (SGD) and
another dataset collected from a real-world dialog application. Experiment
results show that the prompt pool method achieves much higher joint goal
accuracy than the baseline. After combining with a rehearsal buffer, the model
performance can be further improved. | cs.CL | Prompt Pool based Class-Incremental Continual Learning for Dialog State Tracking |
2023-11-17T01:27:01Z | http://arxiv.org/pdf/2311.10267v1 | http://arxiv.org/abs/2311.10267v1 | Xiaorong Wang, Clara Na, Emma Strubell, Sorelle Friedler, Sasha Luccioni | 2023-11-17T01:27:01Z | cs.CL, cs.LG | Despite the popularity of the `pre-train then fine-tune' paradigm in the NLP
community, existing work quantifying energy costs and associated carbon
emissions has largely focused on language model pre-training. Although a single
pre-training run draws substantially more energy than fine-tuning, fine-tuning
is performed more frequently by many more individual actors, and thus must be
accounted for when considering the energy and carbon footprint of NLP. In order
to better characterize the role of fine-tuning in the landscape of energy and
carbon emissions in NLP, we perform a careful empirical study of the
computational costs of fine-tuning across tasks, datasets, hardware
infrastructure and measurement modalities. Our experimental results allow us to
place fine-tuning energy and carbon costs into perspective with respect to
pre-training and inference, and outline recommendations to NLP researchers and
practitioners who wish to improve their fine-tuning energy efficiency. | cs.CL | Energy and Carbon Considerations of Fine-Tuning BERT |
2023-11-17T01:20:08Z | http://arxiv.org/pdf/2311.10266v1 | http://arxiv.org/abs/2311.10266v1 | Ambri Ma, Arnav Kumar, Brett Zeligson | 2023-11-17T01:20:08Z | cs.CL | The training of large language models (LLMs) on extensive, unfiltered corpora
sourced from the internet is a common and advantageous practice. Consequently,
LLMs have learned and inadvertently reproduced various types of biases,
including violent, offensive, and toxic language. However, recent research
shows that generative pretrained transformer (GPT) language models can
recognize their own biases and detect toxicity in generated content, a process
referred to as self-diagnosis. In response, researchers have developed a
decoding algorithm that allows LLMs to self-debias, or reduce their likelihood
of generating harmful text. This study investigates the efficacy of the
diagnosing-debiasing approach in mitigating two additional types of biases:
insults and political bias. These biases are often used interchangeably in
discourse, despite exhibiting potentially dissimilar semantic and syntactic
properties. We aim to contribute to the ongoing effort of investigating the
ethical and social implications of human-AI interaction. | cs.CL | Diagnosing and Debiasing Corpus-Based Political Bias and Insults in GPT2 |
2023-11-16T23:49:55Z | http://arxiv.org/pdf/2311.10236v1 | http://arxiv.org/abs/2311.10236v1 | Maike Züfle, Verna Dankers, Ivan Titov | 2023-11-16T23:49:55Z | cs.CL | With the ever-growing presence of social media platforms comes the increased
spread of harmful content and the need for robust hate speech detection
systems. Such systems easily overfit to specific targets and keywords, and
evaluating them without considering distribution shifts that might occur
between train and test data overestimates their benefit. We challenge hate
speech models via new train-test splits of existing datasets that rely on the
clustering of models' hidden representations. We present two split variants
(Subset-Sum-Split and Closest-Split) that, when applied to two datasets using
four pretrained models, reveal how models catastrophically fail on blind spots
in the latent space. This result generalises when developing a split with one
model and evaluating it on another. Our analysis suggests that there is no
clear surface-level property of the data split that correlates with the
decreased performance, which underscores that task difficulty is not always
humanly interpretable. We recommend incorporating latent feature-based splits
in model development and release two splits via the GenBench benchmark. | cs.CL | Latent Feature-based Data Splits to Improve Generalisation Evaluation: A Hate Speech Detection Case Study |
2023-11-16T22:49:27Z | http://arxiv.org/pdf/2311.10227v1 | http://arxiv.org/abs/2311.10227v1 | Alex Wilf, Sihyun Shawn Lee, Paul Pu Liang, Louis-Philippe Morency | 2023-11-16T22:49:27Z | cs.AI, cs.CL | Human interactions are deeply rooted in the interplay of thoughts, beliefs,
and desires made possible by Theory of Mind (ToM): our cognitive ability to
understand the mental states of ourselves and others. Although ToM may come
naturally to us, emulating it presents a challenge to even the most advanced
Large Language Models (LLMs). Recent improvements to LLMs' reasoning
capabilities from simple yet effective prompting techniques such as
Chain-of-Thought have seen limited applicability to ToM. In this paper, we turn
to the prominent cognitive science theory "Simulation Theory" to bridge this
gap. We introduce SimToM, a novel two-stage prompting framework inspired by
Simulation Theory's notion of perspective-taking. To implement this idea on
current ToM benchmarks, SimToM first filters context based on what the
character in question knows before answering a question about their mental
state. Our approach, which requires no additional training and minimal
prompt-tuning, shows substantial improvement over existing methods, and our
analysis reveals the importance of perspective-taking to Theory-of-Mind
capabilities. Our findings suggest perspective-taking as a promising direction
for future research into improving LLMs' ToM capabilities. | cs.AI | Think Twice: Perspective-Taking Improves Large Language Models' Theory-of-Mind Capabilities |
2023-11-16T22:15:15Z | http://arxiv.org/pdf/2311.10217v2 | http://arxiv.org/abs/2311.10217v2 | Vasilii A. Gromov, Nikita S. Borodin, Asel S. Yerbolova | 2023-11-20T17:08:11Z | cs.CL, cs.AI, math.AT, nlin.CD | The present paper introduces a novel object of study - a language fractal
structure. We hypothesize that a set of embeddings of all $n$-grams of a
natural language constitutes a representative sample of this fractal set. (We
use the term Hailonakea to refer to the sum total of all language fractal
structures, over all $n$). The paper estimates intrinsic (genuine) dimensions
of language fractal structures for the Russian and English languages. To this
end, we employ methods based on (1) topological data analysis and (2) a minimum
spanning tree of a data graph for a cloud of points considered (Steele
theorem). For both languages, for all $n$, the intrinsic dimensions appear to
be non-integer values (typical for fractal sets), close to 9 for both of the
Russian and English language. | cs.CL | A Language and Its Dimensions: Intrinsic Dimensions of Language Fractal Structures |
2023-11-16T22:11:12Z | http://arxiv.org/pdf/2311.10215v1 | http://arxiv.org/abs/2311.10215v1 | Jan Kulveit, Clem von Stengel, Roman Leventov | 2023-11-16T22:11:12Z | cs.CL, cs.AI | Large language models (LLMs) like GPT are often conceptualized as passive
predictors, simulators, or even stochastic parrots. We instead conceptualize
LLMs by drawing on the theory of active inference originating in cognitive
science and neuroscience. We examine similarities and differences between
traditional active inference systems and LLMs, leading to the conclusion that,
currently, LLMs lack a tight feedback loop between acting in the world and
perceiving the impacts of their actions, but otherwise fit in the active
inference paradigm. We list reasons why this loop may soon be closed, and
possible consequences of this including enhanced model self-awareness and the
drive to minimize prediction error by changing the world. | cs.CL | Predictive Minds: LLMs As Atypical Active Inference Agents |
2023-11-16T20:13:24Z | http://arxiv.org/pdf/2311.10181v1 | http://arxiv.org/abs/2311.10181v1 | Yunke He, Xixian Liao, Jialing Liang, Gemma Boleda | 2023-11-16T20:13:24Z | cs.CL | Different speakers often produce different names for the same object or
entity (e.g., "woman" vs. "tourist" for a female tourist). The reasons behind
variation in naming are not well understood. We create a Language and Vision
dataset for Mandarin Chinese that provides an average of 20 names for 1319
naturalistic images, and investigate how familiarity with a given kind of
object relates to the degree of naming variation it triggers across subjects.
We propose that familiarity influences naming variation in two competing ways:
increasing familiarity can either expand vocabulary, leading to higher
variation, or promote convergence on conventional names, thereby reducing
variation. We find evidence for both factors being at play. Our study
illustrates how computational resources can be used to address research
questions in Cognitive Science. | cs.CL | The Impact of Familiarity on Naming Variation: A Study on Object Naming in Mandarin Chinese |
2023-11-16T20:02:44Z | http://arxiv.org/pdf/2311.10174v1 | http://arxiv.org/abs/2311.10174v1 | Shester Gueuwou, Sophie Siake, Colin Leong, Mathias Müller | 2023-11-16T20:02:44Z | cs.CL | Advancements in sign language processing have been hindered by a lack of
sufficient data, impeding progress in recognition, translation, and production
tasks. The absence of comprehensive sign language datasets across the world's
sign languages has widened the gap in this field, resulting in a few sign
languages being studied more than others, making this research area extremely
skewed mostly towards sign languages from high-income countries. In this work
we introduce a new large and highly multilingual dataset for sign language
translation: JWSign. The dataset consists of 2,530 hours of Bible translations
in 98 sign languages, featuring more than 1,500 individual signers. On this
dataset, we report neural machine translation experiments. Apart from bilingual
baseline systems, we also train multilingual systems, including some that take
into account the typological relatedness of signed or spoken languages. Our
experiments highlight that multilingual systems are superior to bilingual
baselines, and that in higher-resource scenarios, clustering language pairs
that are related improves translation quality. | cs.CL | JWSign: A Highly Multilingual Corpus of Bible Translations for more Diversity in Sign Language Processing |
2023-11-16T18:44:22Z | http://arxiv.org/pdf/2311.10085v1 | http://arxiv.org/abs/2311.10085v1 | Fnu Devvrit, Sai Surya Duvvuri, Rohan Anil, Vineet Gupta, Cho-Jui Hsieh, Inderjit Dhillon | 2023-11-16T18:44:22Z | cs.LG, cs.CL, math.OC | Second-order methods hold significant promise for enhancing the convergence
of deep neural network training; however, their large memory and computational
demands have limited their practicality. Thus there is a need for scalable
second-order methods that can efficiently train large models. In this paper, we
introduce the Sparsified Online Newton (SONew) method, a memory-efficient
second-order algorithm that yields a sparsified yet effective preconditioner.
The algorithm emerges from a novel use of the LogDet matrix divergence measure;
we combine it with sparsity constraints to minimize regret in the online convex
optimization framework. Empirically, we test our method on large scale
benchmarks of up to 1B parameters. We achieve up to 30% faster convergence,
3.4% relative improvement in validation performance, and 80% relative
improvement in training loss, in comparison to memory efficient optimizers
including first order methods. Powering the method is a surprising fact --
imposing structured sparsity patterns, like tridiagonal and banded structure,
requires little to no overhead, making it as efficient and parallelizable as
first-order methods. In wall-clock time, tridiagonal SONew is only about 3%
slower per step than first-order methods but gives overall gains due to much
faster convergence. In contrast, one of the state-of-the-art (SOTA)
memory-intensive second-order methods, Shampoo, is unable to scale to large
benchmarks. Additionally, while Shampoo necessitates significant engineering
efforts to scale to large benchmarks, SONew offers a more straightforward
implementation, increasing its practical appeal. SONew code is available at:
https://github.com/devvrit/SONew | cs.LG | A Computationally Efficient Sparsified Online Newton Method |
2023-11-16T18:42:37Z | http://arxiv.org/pdf/2311.10785v1 | http://arxiv.org/abs/2311.10785v1 | Federico Albanese, Daniel Ciolek, Nicolas D'Ippolito | 2023-11-16T18:42:37Z | cs.CL, cs.LG | In the context of information systems, text sanitization techniques are used
to identify and remove sensitive data to comply with security and regulatory
requirements. Even though many methods for privacy preservation have been
proposed, most of them are focused on the detection of entities from specific
domains (e.g., credit card numbers, social security numbers), lacking
generality and requiring customization for each desirable domain. Moreover,
removing words is, in general, a drastic measure, as it can degrade text
coherence and contextual information. Less severe measures include substituting
a word for a safe alternative, yet it can be challenging to automatically find
meaningful substitutions. We present a zero-shot text sanitization technique
that detects and substitutes potentially sensitive information using Large
Language Models. Our evaluation shows that our method excels at protecting
privacy while maintaining text coherence and contextual information, preserving
data utility for downstream tasks. | cs.CL | Text Sanitization Beyond Specific Domains: Zero-Shot Redaction & Substitution with Large Language Models |
2023-11-16T18:38:25Z | http://arxiv.org/pdf/2311.10083v1 | http://arxiv.org/abs/2311.10083v1 | Chung-Ching Chang, William W. Cohen, Yun-Hsuan Sung | 2023-11-16T18:38:25Z | cs.CL | We propose a theoretical framework for formulating language model decoder
algorithms with dynamic programming and information theory. With dynamic
programming, we lift the design of decoder algorithms from the logit space to
the action-state value function space, and show that the decoding algorithms
are consequences of optimizing the action-state value functions. Each component
in the action-state value function space has an information theoretical
interpretation. With the lifting and interpretation, it becomes evident what
the decoder algorithm is optimized for, and hence facilitating the arbitration
of the tradeoffs in sensibleness, diversity, and attribution. | cs.CL | Characterizing Tradeoffs in Language Model Decoding with Informational Interpretations |
2023-11-16T18:37:29Z | http://arxiv.org/pdf/2311.10081v2 | http://arxiv.org/abs/2311.10081v2 | Yangyi Chen, Karan Sikka, Michael Cogswell, Heng Ji, Ajay Divakaran | 2024-03-19T17:51:45Z | cs.CV, cs.CL, cs.LG | We present DRESS, a large vision language model (LVLM) that innovatively
exploits Natural Language feedback (NLF) from Large Language Models to enhance
its alignment and interactions by addressing two key limitations in the
state-of-the-art LVLMs. First, prior LVLMs generally rely only on the
instruction finetuning stage to enhance alignment with human preferences.
Without incorporating extra feedback, they are still prone to generate
unhelpful, hallucinated, or harmful responses. Second, while the visual
instruction tuning data is generally structured in a multi-turn dialogue
format, the connections and dependencies among consecutive conversational turns
are weak. This reduces the capacity for effective multi-turn interactions. To
tackle these, we propose a novel categorization of the NLF into two key types:
critique and refinement. The critique NLF identifies the strengths and
weaknesses of the responses and is used to align the LVLMs with human
preferences. The refinement NLF offers concrete suggestions for improvement and
is adopted to improve the interaction ability of the LVLMs-- which focuses on
LVLMs' ability to refine responses by incorporating feedback in multi-turn
interactions. To address the non-differentiable nature of NLF, we generalize
conditional reinforcement learning for training. Our experimental results
demonstrate that DRESS can generate more helpful (9.76%), honest (11.52%), and
harmless (21.03%) responses, and more effectively learn from feedback during
multi-turn interactions compared to SOTA LVMLs. | cs.CV | DRESS: Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedback |
2023-11-16T18:30:14Z | http://arxiv.org/pdf/2311.10075v1 | http://arxiv.org/abs/2311.10075v1 | Kanhai S. Amin, Linda Mayes, Pavan Khosla, Rushabh Doshi | 2023-11-16T18:30:14Z | cs.CL, cs.AI | Purpose: Enhanced health literacy has been linked to better health outcomes;
however, few interventions have been studied. We investigate whether large
language models (LLMs) can serve as a medium to improve health literacy in
children and other populations.
Methods: We ran 288 conditions using 26 different prompts through
ChatGPT-3.5, Microsoft Bing, and Google Bard. Given constraints imposed by rate
limits, we tested a subset of 150 conditions through ChatGPT-4. The primary
outcome measurements were the reading grade level (RGL) and word counts of
output.
Results: Across all models, output for basic prompts such as "Explain" and
"What is (are)" were at, or exceeded, a 10th-grade RGL. When prompts were
specified to explain conditions from the 1st to 12th RGL, we found that LLMs
had varying abilities to tailor responses based on RGL. ChatGPT-3.5 provided
responses that ranged from the 7th-grade to college freshmen RGL while
ChatGPT-4 outputted responses from the 6th-grade to the college-senior RGL.
Microsoft Bing provided responses from the 9th to 11th RGL while Google Bard
provided responses from the 7th to 10th RGL.
Discussion: ChatGPT-3.5 and ChatGPT-4 did better in achieving lower-grade
level outputs. Meanwhile Bard and Bing tended to consistently produce an RGL
that is at the high school level regardless of prompt. Additionally, Bard's
hesitancy in providing certain outputs indicates a cautious approach towards
health information. LLMs demonstrate promise in enhancing health communication,
but future research should verify the accuracy and effectiveness of such tools
in this context.
Implications: LLMs face challenges in crafting outputs below a sixth-grade
reading level. However, their capability to modify outputs above this threshold
provides a potential mechanism to improve health literacy and communication in
a pediatric population and beyond. | cs.CL | ChatGPT-3.5, ChatGPT-4, Google Bard, and Microsoft Bing to Improve Health Literacy and Communication in Pediatric Populations and Beyond |
2023-11-16T17:52:21Z | http://arxiv.org/pdf/2311.10057v3 | http://arxiv.org/abs/2311.10057v3 | Ilaria Manco, Benno Weck, SeungHeon Doh, Minz Won, Yixiao Zhang, Dmitry Bogdanov, Yusong Wu, Ke Chen, Philip Tovstogan, Emmanouil Benetos, Elio Quinton, György Fazekas, Juhan Nam | 2023-11-22T21:22:11Z | cs.SD, cs.AI, cs.CL, eess.AS | We introduce the Song Describer dataset (SDD), a new crowdsourced corpus of
high-quality audio-caption pairs, designed for the evaluation of
music-and-language models. The dataset consists of 1.1k human-written natural
language descriptions of 706 music recordings, all publicly accessible and
released under Creative Common licenses. To showcase the use of our dataset, we
benchmark popular models on three key music-and-language tasks (music
captioning, text-to-music generation and music-language retrieval). Our
experiments highlight the importance of cross-dataset evaluation and offer
insights into how researchers can use SDD to gain a broader understanding of
model performance. | cs.SD | The Song Describer Dataset: a Corpus of Audio Captions for Music-and-Language Evaluation |
2023-11-16T17:48:55Z | http://arxiv.org/pdf/2311.10054v1 | http://arxiv.org/abs/2311.10054v1 | Mingqian Zheng, Jiaxin Pei, David Jurgens | 2023-11-16T17:48:55Z | cs.CL, cs.AI, cs.CY, cs.HC, cs.LG | Prompting serves as the major way humans interact with Large Language Models
(LLM). Commercial AI systems commonly define the role of the LLM in system
prompts. For example, ChatGPT uses "You are a helpful assistant" as part of the
default system prompt. But is "a helpful assistant" the best role for LLMs? In
this study, we present a systematic evaluation of how social roles in system
prompts affect model performance. We curate a list of 162 roles covering 6
types of interpersonal relationships and 8 types of occupations. Through
extensive analysis of 3 popular LLMs and 2457 questions, we show that adding
interpersonal roles in prompts consistently improves the models' performance
over a range of questions. Moreover, while we find that using gender-neutral
roles and specifying the role as the audience leads to better performances,
predicting which role leads to the best performance remains a challenging task,
and that frequency, similarity, and perplexity do not fully explain the effect
of social roles on model performances. Our results can help inform the design
of system prompts for AI systems. Code and data are available at
https://github.com/Jiaxin-Pei/Prompting-with-Social-Roles. | cs.CL | Is "A Helpful Assistant" the Best Role for Large Language Models? A Systematic Evaluation of Social Roles in System Prompts |
2023-11-16T16:09:43Z | http://arxiv.org/pdf/2311.09993v1 | http://arxiv.org/abs/2311.09993v1 | Sagi Pendzel, Tomer Wullach, Amir Adler, Einat Minkov | 2023-11-16T16:09:43Z | cs.CL, cs.AI | Automatic hate speech detection using deep neural models is hampered by the
scarcity of labeled datasets, leading to poor generalization. To mitigate this
problem, generative AI has been utilized to generate large amounts of synthetic
hate speech sequences from available labeled examples, leveraging the generated
data in finetuning large pre-trained language models (LLMs). In this chapter,
we provide a review of relevant methods, experimental setups and evaluation of
this approach. In addition to general LLMs, such as BERT, RoBERTa and ALBERT,
we apply and evaluate the impact of train set augmentation with generated data
using LLMs that have been already adapted for hate detection, including
RoBERTa-Toxicity, HateBERT, HateXplain, ToxDect, and ToxiGen. An empirical
study corroborates our previous findings, showing that this approach improves
hate speech generalization, boosting recall performance across data
distributions. In addition, we explore and compare the performance of the
finetuned LLMs with zero-shot hate detection using a GPT-3.5 model. Our results
demonstrate that while better generalization is achieved using the GPT-3.5
model, it achieves mediocre recall and low precision on most datasets. It is an
open question whether the sensitivity of models such as GPT-3.5, and onward,
can be improved using similar techniques of text generation. | cs.CL | Generative AI for Hate Speech Detection: Evaluation and Findings |
2023-11-16T15:57:58Z | http://arxiv.org/pdf/2311.10784v1 | http://arxiv.org/abs/2311.10784v1 | Sabrine Amri, Henri-Cedric Mputu Boleilanga, Esma Aïmeur | 2023-11-16T15:57:58Z | cs.CL, cs.AI, cs.LG, cs.SI | ExFake is an explainable fake news detection system based on content and
context-level information. It is concerned with the veracity analysis of online
posts based on their content, social context (i.e., online users' credibility
and historical behaviour), and data coming from trusted entities such as
fact-checking websites and named entities. Unlike state-of-the-art systems, an
Explainable AI (XAI) assistant is also adopted to help online social networks
(OSN) users develop good reflexes when faced with any doubted information that
spreads on social networks. The trustworthiness of OSN users is also addressed
by assigning a credibility score to OSN users, as OSN users are one of the main
culprits for spreading fake news. Experimental analysis on a real-world dataset
demonstrates that ExFake significantly outperforms other baseline methods for
fake news detection. | cs.CL | ExFake: Towards an Explainable Fake News Detection Based on Content and Social Context Information |
2023-11-16T15:52:24Z | http://arxiv.org/pdf/2311.09979v1 | http://arxiv.org/abs/2311.09979v1 | Tomoyuki Yamakami | 2023-11-16T15:52:24Z | cs.FL, cs.CC, cs.CL | Nonuniform families of polynomial-size finite automata, which are series of
indexed finite automata having polynomially many inner states, are used in the
past literature to solve nonuniform families of promise decision problems.
Among such nonuniform families of finite automata, we focus our attention, in
particular, on the variants of nondeterministic finite automata, which have at
most "one" (unambiguous), "polynomially many" (few) accepting computation
paths, or unambiguous/few computation paths leading to each fixed
configuration. When such machines are limited to make only one-way head moves,
we can prove with no unproven hardness assumptions that some of these variants
are different in computational power from each other. As for two-way machines
restricted to instances of polynomially-bounded length, families of two-way
polynomial-size nondeterministic finite automata are equivalent in power to
families of polynomial-size unambiguous finite automata. | cs.FL | Unambiguity and Fewness for Nonuniform Families of Polynomial-Size Nondeterministic Finite Automata |
2023-11-16T15:01:48Z | http://arxiv.org/pdf/2311.09948v2 | http://arxiv.org/abs/2311.09948v2 | Yao Qiang, Xiangyu Zhou, Dongxiao Zhu | 2024-06-15T18:54:54Z | cs.LG, cs.CL, cs.CR | In-context learning (ICL) has emerged as a powerful paradigm leveraging LLMs
for specific downstream tasks by utilizing labeled examples as demonstrations
(demos) in the precondition prompts. Despite its promising performance, ICL
suffers from instability with the choice and arrangement of examples.
Additionally, crafted adversarial attacks pose a notable threat to the
robustness of ICL. However, existing attacks are either easy to detect, rely on
external models, or lack specificity towards ICL. This work introduces a novel
transferable attack against ICL to address these issues, aiming to hijack LLMs
to generate the target response or jailbreak. Our hijacking attack leverages a
gradient-based prompt search method to learn and append imperceptible
adversarial suffixes to the in-context demos without directly contaminating the
user queries. Comprehensive experimental results across different generation
and jailbreaking tasks highlight the effectiveness of our hijacking attack,
resulting in distracted attention towards adversarial tokens and consequently
leading to unwanted target outputs. We also propose a defense strategy against
hijacking attacks through the use of extra clean demos, which enhances the
robustness of LLMs during ICL. Broadly, this work reveals the significant
security vulnerabilities of LLMs and emphasizes the necessity for in-depth
studies on their robustness. | cs.LG | Hijacking Large Language Models via Adversarial In-Context Learning |
2023-11-16T14:56:09Z | http://arxiv.org/pdf/2311.09945v1 | http://arxiv.org/abs/2311.09945v1 | Qirui Tang, Wenkang Jiang, Yihua Du, Lei Lin | 2023-11-16T14:56:09Z | cs.CY, cs.CL | In social media networks, users produce a large amount of text content
anytime, providing researchers with a valuable approach to digging for
personality-related information. Personality detection based on user-generated
texts is a universal method that can be used to build user portraits. The
presence of noise in social media texts hinders personality detection. However,
previous studies have not fully addressed this challenge. Inspired by the
scanning reading technique, we propose an attention-based information
extraction mechanism (AIEM) for long texts, which is applied to quickly locate
valuable pieces of information, and focus more attention on the deep semantics
of key pieces. Then, we provide a novel attention-based denoising framework
(ADF) for personality detection tasks and achieve state-of-the-art performance
on two commonly used datasets. Notably, we obtain an average accuracy
improvement of 10.2% on the gold standard Twitter-Myers-Briggs Type Indicator
(Twitter-MBTI) dataset. We made our code publicly available on GitHub. We shed
light on how AIEM works to magnify personality-related signals. | cs.CY | An Attention-Based Denoising Framework for Personality Detection in Social Media Texts |
2023-11-16T14:18:24Z | http://arxiv.org/pdf/2311.10782v1 | http://arxiv.org/abs/2311.10782v1 | Sayan Putatunda, Anwesha Bhowmik, Girish Thiruvenkadam, Rahul Ghosh | 2023-11-16T14:18:24Z | cs.CL, cs.IR, cs.LG | According to the literature, Product reviews are an important source of
information for customers to support their buying decision. Product reviews
improve customer trust and loyalty. Reviews help customers in understanding
what other customers think about a particular product and helps in driving
purchase decisions. Therefore, for an e-commerce platform it is important to
understand the sentiments in customer reviews to understand their products and
services, and it also allows them to potentially create positive consumer
interaction as well as long lasting relationships. Reviews also provide
innovative ways to market the products for an ecommerce company. One such
approach is Nudge Marketing. Nudge marketing is a subtle way for an ecommerce
company to help their customers make better decisions without hesitation. | cs.CL | A BERT based Ensemble Approach for Sentiment Classification of Customer Reviews and its Application to Nudge Marketing in e-Commerce |
2023-11-16T13:37:21Z | http://arxiv.org/pdf/2311.09889v5 | http://arxiv.org/abs/2311.09889v5 | Ziyi Ye, Qingyao Ai, Yiqun Liu, Maarten de Rijke, Min Zhang, Christina Lioma, Tuukka Ruotsalo | 2024-03-11T11:05:21Z | cs.CL | Generating human language through non-invasive brain-computer interfaces
(BCIs) has the potential to unlock many applications, such as serving disabled
patients and improving communication. Currently, however, generating language
via BCIs has been previously successful only within a classification setup for
selecting pre-generated sentence continuation candidates with the most likely
cortical semantic representation. Inspired by recent research that revealed
associations between the brain and the large computational language models, we
propose a generative language BCI that utilizes the capacity of a large
language model (LLM) jointly with a semantic brain decoder to directly generate
language from functional magnetic resonance imaging (fMRI) input. The proposed
model can generate coherent language sequences aligned with the semantic
content of visual or auditory language stimuli perceived, without prior
knowledge of any pre-generated candidates. We compare the language generated
from the presented model with a random control, pre-generated language
selection approach, and a standard LLM, which generates common coherent text
solely based on the next word likelihood according to statistical language
training data. The proposed model is found to generate language that is more
aligned with semantic stimulus in response to which brain input is sampled. Our
findings demonstrate the potential and feasibility of employing BCIs in direct
language generation. | cs.CL | Language Generation from Brain Recordings |
2023-11-16T12:45:41Z | http://arxiv.org/pdf/2311.09862v2 | http://arxiv.org/abs/2311.09862v2 | Debarati Das, Ishaan Gupta, Jaideep Srivastava, Dongyeop Kang | 2024-03-13T23:32:32Z | cs.CL, cs.SI | Our research integrates graph data with Large Language Models (LLMs), which,
despite their advancements in various fields using large text corpora, face
limitations in encoding entire graphs due to context size constraints. This
paper introduces a new approach to encoding a graph with diverse modalities,
such as text, image, and motif, coupled with prompts to approximate a graph's
global connectivity, thereby enhancing LLMs' efficiency in processing complex
graph structures. The study also presents GraphTMI, a novel benchmark for
evaluating LLMs in graph structure analysis, focusing on homophily, motif
presence, and graph difficulty. Key findings indicate that the image modality,
especially with vision-language models like GPT-4V, is superior to text in
balancing token limits and preserving essential information and outperforms
prior graph neural net (GNN) encoders. Furthermore, the research assesses how
various factors affect the performance of each encoding modality and outlines
the existing challenges and potential future developments for LLMs in graph
understanding and reasoning tasks. All data will be publicly available upon
acceptance. | cs.CL | Which Modality should I use -- Text, Motif, or Image? : Understanding Graphs with Large Language Models |
2023-11-16T12:43:18Z | http://arxiv.org/pdf/2311.09861v4 | http://arxiv.org/abs/2311.09861v4 | Junlei Zhang, Hongliang He, Nirui Song, Zhanchao Zhou, Shuyuan He, Shuai Zhang, Huachuan Qiu, Anqi Li, Yong Dai, Lizhi Ma, Zhenzhong Lan | 2024-06-16T11:33:03Z | cs.CL, cs.AI | The critical field of psychology necessitates a comprehensive benchmark to
enhance the evaluation and development of domain-specific Large Language Models
(LLMs). Existing MMLU-type benchmarks, such as C-EVAL and CMMLU, include
psychology-related subjects, but their limited number of questions and lack of
systematic concept sampling strategies mean they cannot cover the concepts
required in psychology. Consequently, despite their broad subject coverage,
these benchmarks lack the necessary depth in the psychology domain, making them
inadequate as psychology-specific evaluation suite. To address this issue, this
paper presents ConceptPsy, designed to evaluate Chinese complex reasoning and
knowledge abilities in psychology. ConceptPsy includes 12 core subjects and
1383 manually collected concepts. Specifically, we prompt GPT-4 to generate
questions for each concept using carefully designed diverse prompts and hire
professional psychologists to review these questions. To help to understand the
fine-grained performances and enhance the weaknesses, we annotate each question
with a chapter label and provide chapter-wise accuracy. Based on ConceptPsy, we
evaluate a broad range of LLMs. We observe that, although some LLMs achieve
similar accuracies on overall performances, they exhibit significant
performance variations across different psychology concepts, even when they are
models from the same series. We hope our work can facilitate the development of
LLMs in the field of psychology. | cs.CL | ConceptPsy:A Benchmark Suite with Conceptual Comprehensiveness in Psychology |
2023-11-16T12:43:02Z | http://arxiv.org/pdf/2311.09860v1 | http://arxiv.org/abs/2311.09860v1 | Wolfgang Otto, Matthäus Zloch, Lu Gan, Saurav Karmakar, Stefan Dietze | 2023-11-16T12:43:02Z | cs.CL | Named Entity Recognition (NER) models play a crucial role in various NLP
tasks, including information extraction (IE) and text understanding. In
academic writing, references to machine learning models and datasets are
fundamental components of various computer science publications and necessitate
accurate models for identification. Despite the advancements in NER, existing
ground truth datasets do not treat fine-grained types like ML model and model
architecture as separate entity types, and consequently, baseline models cannot
recognize them as such. In this paper, we release a corpus of 100 manually
annotated full-text scientific publications and a first baseline model for 10
entity types centered around ML models and datasets. In order to provide a
nuanced understanding of how ML models and datasets are mentioned and utilized,
our dataset also contains annotations for informal mentions like "our
BERT-based model" or "an image CNN". You can find the ground truth dataset and
code to replicate model training at https://data.gesis.org/gsap/gsap-ner. | cs.CL | GSAP-NER: A Novel Task, Corpus, and Baseline for Scholarly Entity Extraction Focused on Machine Learning Models and Datasets |
2023-11-16T12:13:49Z | http://arxiv.org/pdf/2311.09841v1 | http://arxiv.org/abs/2311.09841v1 | Tilahun Abedissa Taffa, Ricardo Usbeck | 2023-11-16T12:13:49Z | cs.CL, cs.AI, cs.DB, cs.LG | This paper presents a scholarly Knowledge Graph Question Answering (KGQA)
that answers bibliographic natural language questions by leveraging a large
language model (LLM) in a few-shot manner. The model initially identifies the
top-n similar training questions related to a given test question via a
BERT-based sentence encoder and retrieves their corresponding SPARQL. Using the
top-n similar question-SPARQL pairs as an example and the test question creates
a prompt. Then pass the prompt to the LLM and generate a SPARQL. Finally, runs
the SPARQL against the underlying KG - ORKG (Open Research KG) endpoint and
returns an answer. Our system achieves an F1 score of 99.0%, on SciQA - one of
the Scholarly-QALD-23 challenge benchmarks. | cs.CL | Leveraging LLMs in Scholarly Knowledge Graph Question Answering |
2023-11-16T12:05:23Z | http://arxiv.org/pdf/2311.09836v1 | http://arxiv.org/abs/2311.09836v1 | Joseph J. Peper, Wenzhao Qiu, Lu Wang | 2023-11-16T12:05:23Z | cs.CL, cs.AI | We investigate pre-training techniques for abstractive multi-document
summarization (MDS), which is much less studied than summarizing single
documents. Though recent work has demonstrated the effectiveness of
highlighting information salience for pre-training strategy design, it
struggles to generate abstractive and reflective summaries, which are critical
properties for MDS. To this end, we present PELMS, a pre-trained model that
uses objectives based on semantic coherence heuristics and faithfulness
constraints with un-labeled multi-document inputs, to promote the generation of
concise, fluent, and faithful summaries. To support the training of PELMS, we
compile MultiPT, a multi-document pre-training corpus containing over 93
million documents to form more than 3 million unlabeled topic-centric document
clusters, covering diverse genres such as product reviews, news, and general
knowledge. We perform extensive evaluation of PELMS in low-shot settings on a
wide range of MDS datasets. Our approach consistently outperforms competitive
comparisons with respect to overall informativeness, abstractiveness,
coherence, and faithfulness. | cs.CL | PELMS: Pre-training for Effective Low-Shot Multi-Document Summarization |
2023-11-16T12:03:21Z | http://arxiv.org/pdf/2311.09835v4 | http://arxiv.org/abs/2311.09835v4 | Xiangru Tang, Yuliang Liu, Zefan Cai, Yanjun Shao, Junjie Lu, Yichi Zhang, Zexuan Deng, Helan Hu, Kaikai An, Ruijun Huang, Shuzheng Si, Sheng Chen, Haozhe Zhao, Liang Chen, Yan Wang, Tianyu Liu, Zhiwei Jiang, Baobao Chang, Yin Fang, Yujia Qin, Wangchunshu Zhou, Yilun Zhao, Arman Cohan, Mark Gerstein | 2024-06-18T12:49:41Z | cs.CL, cs.AI | Despite Large Language Models (LLMs) like GPT-4 achieving impressive results
in function-level code generation, they struggle with repository-scale code
understanding (e.g., coming up with the right arguments for calling routines),
requiring a deeper comprehension of complex file interactions. Also, recently,
people have developed LLM agents that attempt to interact with repository code
(e.g., compiling and evaluating its execution), prompting the need to evaluate
their performance. These gaps have motivated our development of ML-Bench, a
benchmark rooted in real-world programming applications that leverage existing
code repositories to perform tasks. Addressing the need for LLMs to interpret
long code contexts and translate instructions into precise, executable scripts,
ML-Bench encompasses annotated 9,641 examples across 18 GitHub repositories,
challenging LLMs to accommodate user-specified arguments and documentation
intricacies effectively. To evaluate both LLMs and AI agents, two setups are
employed: ML-LLM-Bench for assessing LLMs' text-to-code conversion within a
predefined deployment environment, and ML-Agent-Bench for testing autonomous
agents in an end-to-end task execution within a Linux sandbox environment. Our
findings indicate that while GPT-4o leads with a Pass@5 rate surpassing 50%,
there remains significant scope for improvement, highlighted by issues such as
hallucinated outputs and difficulties with bash script generation. Notably, in
the more demanding ML-Agent-Bench, GPT-4o achieves a 76.47% success rate,
reflecting the efficacy of iterative action and feedback in complex task
resolution. Our code, dataset, and models are available at
https://github.com/gersteinlab/ML-bench. | cs.CL | ML-Bench: Evaluating Large Language Models and Agents for Machine Learning Tasks on Repository-Level Code |
2023-11-16T12:01:19Z | http://arxiv.org/pdf/2311.09834v1 | http://arxiv.org/abs/2311.09834v1 | Sarah Masud, Mohammad Aflah Khan, Md. Shad Akhtar, Tanmoy Chakraborty | 2023-11-16T12:01:19Z | cs.CL | As hate speech continues to proliferate on the web, it is becoming
increasingly important to develop computational methods to mitigate it.
Reactively, using black-box models to identify hateful content can perplex
users as to why their posts were automatically flagged as hateful. On the other
hand, proactive mitigation can be achieved by suggesting rephrasing before a
post is made public. However, both mitigation techniques require information
about which part of a post contains the hateful aspect, i.e., what spans within
a text are responsible for conveying hate. Better detection of such spans can
significantly reduce explicitly hateful content on the web. To further
contribute to this research area, we organized HateNorm at HASOC-FIRE 2023,
focusing on explicit span detection in English Tweets. A total of 12 teams
participated in the competition, with the highest macro-F1 observed at 0.58. | cs.CL | Overview of the HASOC Subtrack at FIRE 2023: Identification of Tokens Contributing to Explicit Hate in English by Span Detection |
2023-11-16T11:58:31Z | http://arxiv.org/pdf/2311.09832v3 | http://arxiv.org/abs/2311.09832v3 | Liang Chen, Yatao Bian, Yang Deng, Deng Cai, Shuaiyi Li, Peilin Zhao, Kam-fai Wong | 2024-06-06T13:17:43Z | cs.CL | Text watermarking has emerged as a pivotal technique for identifying
machine-generated text. However, existing methods often rely on arbitrary
vocabulary partitioning during decoding to embed watermarks, which compromises
the availability of suitable tokens and significantly degrades the quality of
responses. This study assesses the impact of watermarking on different
capabilities of large language models (LLMs) from a cognitive science lens. Our
finding highlights a significant disparity; knowledge recall and logical
reasoning are more adversely affected than language generation. These results
suggest a more profound effect of watermarking on LLMs than previously
understood. To address these challenges, we introduce Watermarking with Mutual
Exclusion (WatME), a novel approach leveraging linguistic prior knowledge of
inherent lexical redundancy in LLM vocabularies to seamlessly integrate
watermarks. Specifically, WatME dynamically optimizes token usage during the
decoding process by applying a mutually exclusive rule to the identified
lexical redundancies. This strategy effectively prevents the unavailability of
appropriate tokens and preserves the expressive power of LLMs. We provide both
theoretical analysis and empirical evidence showing that WatME effectively
preserves the diverse capabilities of LLMs while ensuring watermark
detectability. | cs.CL | WatME: Towards Lossless Watermarking Through Lexical Redundancy |
2023-11-16T11:55:27Z | http://arxiv.org/pdf/2311.09830v2 | http://arxiv.org/abs/2311.09830v2 | Katharina Stein, Daniel Fišer, Jörg Hoffmann, Alexander Koller | 2024-02-09T09:48:41Z | cs.AI, cs.CL | LLMs are being increasingly used for planning-style tasks, but their
capabilities for planning and reasoning are poorly understood. We present
AutoPlanBench, a novel method for automatically converting planning benchmarks
written in PDDL into textual descriptions and offer a benchmark dataset created
with our method. We show that while the best LLM planners do well on some
planning tasks, others remain out of reach of current methods. | cs.AI | AutoPlanBench: Automatically generating benchmarks for LLM planners from PDDL |
2023-11-16T11:53:31Z | http://arxiv.org/pdf/2311.09829v1 | http://arxiv.org/abs/2311.09829v1 | Yimin Jing, Renren Jin, Jiahao Hu, Huishi Qiu, Xiaohua Wang, Peng Wang, Deyi Xiong | 2023-11-16T11:53:31Z | cs.CL | The effective assessment of the instruction-following ability of large
language models (LLMs) is of paramount importance. A model that cannot adhere
to human instructions might be not able to provide reliable and helpful
responses. In pursuit of this goal, various benchmarks have been constructed to
evaluate the instruction-following capacity of these models. However, these
benchmarks are limited to a single language and are constructed using automated
approaches, which restricts their applicability and the quality of the test
examples they contain. To bridge this gap, we introduce the FollowEval
benchmark in this paper. This benchmark is composed of instances in both
English and Chinese, and all test examples are crafted by human experts.
Furthermore, the FollowEval benchmark is designed to assess LLMs across five
critical dimensions of instruction following: string manipulation, commonsense
reasoning, logical reasoning, spatial reasoning, and response constraints. To
enhance the complexity and present a sufficient challenge, each test example is
designed to evaluate more than one dimension. We have evaluated various LLMs
using the FollowEval benchmark and found that their performance significantly
lags behind that of humans. This highlights the considerable room for
improvement in the instruction-following ability of these models. | cs.CL | FollowEval: A Multi-Dimensional Benchmark for Assessing the Instruction-Following Capability of Large Language Models |
2023-11-16T11:52:52Z | http://arxiv.org/pdf/2311.09828v3 | http://arxiv.org/abs/2311.09828v3 | Jiayi Wang, David Ifeoluwa Adelani, Sweta Agrawal, Marek Masiak, Ricardo Rei, Eleftheria Briakou, Marine Carpuat, Xuanli He, Sofia Bourhim, Andiswa Bukula, Muhidin Mohamed, Temitayo Olatoye, Tosin Adewumi, Hamam Mokayed, Christine Mwase, Wangui Kimotho, Foutse Yuehgoh, Anuoluwapo Aremu, Jessica Ojo, Shamsuddeen Hassan Muhammad, Salomey Osei, Abdul-Hakeem Omotayo, Chiamaka Chukwuneke, Perez Ogayo, Oumaima Hourrane, Salma El Anigri, Lolwethu Ndolela, Thabiso Mangwana, Shafie Abdi Mohamed, Ayinde Hassan, Oluwabusayo Olufunke Awoyomi, Lama Alkhaled, Sana Al-Azzawi, Naome A. Etori, Millicent Ochieng, Clemencia Siro, Samuel Njoroge, Eric Muchiri, Wangari Kimotho, Lyse Naomi Wamba Momo, Daud Abolade, Simbiat Ajao, Iyanuoluwa Shode, Ricky Macharm, Ruqayya Nasir Iro, Saheed S. Abdullahi, Stephen E. Moore, Bernard Opoku, Zainab Akinjobi, Abeeb Afolabi, Nnaemeka Obiefuna, Onyekachi Raphael Ogbu, Sam Brian, Verrah Akinyi Otiende, Chinedu Emmanuel Mbonu, Sakayo Toadoum Sari, Yao Lu, Pontus Stenetorp | 2024-04-23T18:22:05Z | cs.CL | Despite the recent progress on scaling multilingual machine translation (MT)
to several under-resourced African languages, accurately measuring this
progress remains challenging, since evaluation is often performed on n-gram
matching metrics such as BLEU, which typically show a weaker correlation with
human judgments. Learned metrics such as COMET have higher correlation;
however, the lack of evaluation data with human ratings for under-resourced
languages, complexity of annotation guidelines like Multidimensional Quality
Metrics (MQM), and limited language coverage of multilingual encoders have
hampered their applicability to African languages. In this paper, we address
these challenges by creating high-quality human evaluation data with simplified
MQM guidelines for error detection and direct assessment (DA) scoring for 13
typologically diverse African languages. Furthermore, we develop AfriCOMET:
COMET evaluation metrics for African languages by leveraging DA data from
well-resourced languages and an African-centric multilingual encoder
(AfroXLM-R) to create the state-of-the-art MT evaluation metrics for African
languages with respect to Spearman-rank correlation with human judgments
(0.441). | cs.CL | AfriMTE and AfriCOMET: Enhancing COMET to Embrace Under-resourced African Languages |
2023-11-16T11:52:22Z | http://arxiv.org/pdf/2311.09827v2 | http://arxiv.org/abs/2311.09827v2 | Nan Xu, Fei Wang, Ben Zhou, Bang Zheng Li, Chaowei Xiao, Muhao Chen | 2024-02-29T08:20:07Z | cs.CL | While large language models (LLMs) have demonstrated increasing power, they
have also given rise to a wide range of harmful behaviors. As representatives,
jailbreak attacks can provoke harmful or unethical responses from LLMs, even
after safety alignment. In this paper, we investigate a novel category of
jailbreak attacks specifically designed to target the cognitive structure and
processes of LLMs. Specifically, we analyze the safety vulnerability of LLMs in
the face of (1) multilingual cognitive overload, (2) veiled expression, and (3)
effect-to-cause reasoning. Different from previous jailbreak attacks, our
proposed cognitive overload is a black-box attack with no need for knowledge of
model architecture or access to model weights. Experiments conducted on
AdvBench and MasterKey reveal that various LLMs, including both popular
open-source model Llama 2 and the proprietary model ChatGPT, can be compromised
through cognitive overload. Motivated by cognitive psychology work on managing
cognitive load, we further investigate defending cognitive overload attack from
two perspectives. Empirical studies show that our cognitive overload from three
perspectives can jailbreak all studied LLMs successfully, while existing
defense strategies can hardly mitigate the caused malicious uses effectively. | cs.CL | Cognitive Overload: Jailbreaking Large Language Models with Overloaded Logical Thinking |
2023-11-16T11:51:13Z | http://arxiv.org/pdf/2311.09825v1 | http://arxiv.org/abs/2311.09825v1 | Yuxuan Lu, Bingsheng Yao, Shao Zhang, Yun Wang, Peng Zhang, Tun Lu, Toby Jia-Jun Li, Dakuo Wang | 2023-11-16T11:51:13Z | cs.CL | Large Language Models (LLMs) have demonstrated considerable advances, and
several claims have been made about their exceeding human performance. However,
in real-world tasks, domain knowledge is often required. Low-resource learning
methods like Active Learning (AL) have been proposed to tackle the cost of
domain expert annotation, raising this question: Can LLMs surpass compact
models trained with expert annotations in domain-specific tasks? In this work,
we conduct an empirical experiment on four datasets from three different
domains comparing SOTA LLMs with small models trained on expert annotations
with AL. We found that small models can outperform GPT-3.5 with a few hundreds
of labeled data, and they achieve higher or similar performance with GPT-4
despite that they are hundreds time smaller. Based on these findings, we posit
that LLM predictions can be used as a warmup method in real-world applications
and human experts remain indispensable in tasks involving data annotation
driven by domain-specific knowledge. | cs.CL | Human Still Wins over LLM: An Empirical Study of Active Learning on Domain-Specific Annotation Tasks |
2023-11-16T11:49:29Z | http://arxiv.org/pdf/2311.09821v1 | http://arxiv.org/abs/2311.09821v1 | Qingyu Tan, Hwee Tou Ng, Lidong Bing | 2023-11-16T11:49:29Z | cs.CL | Knowledge in the real world is being updated constantly. However, it is
costly to frequently update large language models (LLMs). Therefore, it is
crucial for LLMs to understand the concept of temporal knowledge. However,
prior works on temporal question answering did not emphasize multi-answer and
multi-hop types of temporal reasoning. In this paper, we propose a complex
temporal question-answering (QA) dataset Complex-TR that focuses on
multi-answer and multi-hop temporal reasoning. Besides, we also propose a novel
data augmentation strategy to improve the complex temporal reasoning capability
and robustness of LLMs. We conducted experiments on multiple temporal QA
datasets. Experimental results show that our method is able to improve LLMs'
performance on temporal QA benchmarks by significant margins. | cs.CL | Towards Robust Temporal Reasoning of Large Language Models via a Multi-Hop QA Dataset and Pseudo-Instruction Tuning |
2023-11-16T11:48:17Z | http://arxiv.org/pdf/2311.09818v2 | http://arxiv.org/abs/2311.09818v2 | Shicheng Liu, Jialiang Xu, Wesley Tjangnaka, Sina J. Semnani, Chen Jie Yu, Monica S. Lam | 2024-03-13T17:07:02Z | cs.CL, cs.PL | While most conversational agents are grounded on either free-text or
structured knowledge, many knowledge corpora consist of hybrid sources. This
paper presents the first conversational agent that supports the full generality
of hybrid data access for large knowledge corpora, through a language we
developed called SUQL (Structured and Unstructured Query Language).
Specifically, SUQL extends SQL with free-text primitives (summary and answer),
so information retrieval can be composed with structured data accesses
arbitrarily in a formal, succinct, precise, and interpretable notation. With
SUQL, we propose the first semantic parser, an LLM with in-context learning,
that can handle hybrid data sources.
Our in-context learning-based approach, when applied to the HybridQA dataset,
comes within 8.9% exact match and 7.1% F1 of the SOTA, which was trained on 62K
data samples. More significantly, unlike previous approaches, our technique is
applicable to large databases and free-text corpora. We introduce a dataset
consisting of crowdsourced questions and conversations on Yelp, a large, real
restaurant knowledge base with structured and unstructured data. We show that
our few-shot conversational agent based on SUQL finds an entity satisfying all
user requirements 90.3% of the time, compared to 63.4% for a baseline based on
linearization. | cs.CL | SUQL: Conversational Search over Structured and Unstructured Data with Large Language Models |
2023-11-16T11:47:58Z | http://arxiv.org/pdf/2311.10537v4 | http://arxiv.org/abs/2311.10537v4 | Xiangru Tang, Anni Zou, Zhuosheng Zhang, Ziming Li, Yilun Zhao, Xingyao Zhang, Arman Cohan, Mark Gerstein | 2024-06-04T23:47:43Z | cs.CL, cs.AI | Large language models (LLMs), despite their remarkable progress across
various general domains, encounter significant barriers in medicine and
healthcare. This field faces unique challenges such as domain-specific
terminologies and reasoning over specialized knowledge. To address these
issues, we propose MedAgents, a novel multi-disciplinary collaboration
framework for the medical domain. MedAgents leverages LLM-based agents in a
role-playing setting that participate in a collaborative multi-round
discussion, thereby enhancing LLM proficiency and reasoning capabilities. This
training-free framework encompasses five critical steps: gathering domain
experts, proposing individual analyses, summarising these analyses into a
report, iterating over discussions until a consensus is reached, and ultimately
making a decision. Our work focuses on the zero-shot setting, which is
applicable in real-world scenarios. Experimental results on nine datasets
(MedQA, MedMCQA, PubMedQA, and six subtasks from MMLU) establish that our
proposed MedAgents framework excels at mining and harnessing the medical
expertise within LLMs, as well as extending its reasoning abilities. Our code
can be found at https://github.com/gersteinlab/MedAgents. | cs.CL | MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning |
2023-11-16T11:44:58Z | http://arxiv.org/pdf/2311.09816v1 | http://arxiv.org/abs/2311.09816v1 | Anirudh Ajith, Sameer Singh, Danish Pruthi | 2023-11-16T11:44:58Z | cs.CL, cs.AI, cs.CR, cs.LG | Amidst growing concerns of large language models (LLMs) being misused for
generating misinformation or completing homework assignments, watermarking has
emerged as an effective solution for distinguishing human-written and
LLM-generated text. A prominent watermarking strategy is to embed a signal into
generated text by upsampling a (pseudorandomly-chosen) subset of tokens at
every generation step. Although this signal is imperceptible to a human reader,
it is detectable through statistical testing. However, implanting such signals
alters the model's output distribution and can have unintended effects when
watermarked LLMs are used for downstream applications. In this work, we
evaluate the performance of watermarked LLMs on a diverse suite of tasks,
including text classification, textual entailment, reasoning, question
answering, translation, summarization, and language modeling. We find that
watermarking has negligible impact on the performance of tasks posed as k-class
classification problems in the average case. However, the accuracy can plummet
to that of a random classifier for some scenarios (that occur with
non-negligible probability). Tasks that are cast as multiple-choice questions
and short-form generation are surprisingly unaffected by watermarking. For
long-form generation tasks, including summarization and translation, we see a
drop of 15-20% in the performance due to watermarking. Our findings highlight
the trade-offs that users should be cognizant of when using watermarked models,
and point to cases where future research could improve existing trade-offs. | cs.CL | Performance Trade-offs of Watermarking Large Language Models |
2023-11-16T11:37:54Z | http://arxiv.org/pdf/2311.09812v2 | http://arxiv.org/abs/2311.09812v2 | Maram Hasanain, Fatema Ahmed, Firoj Alam | 2024-01-14T06:32:09Z | cs.CL, 68T50, F.2.2; I.2.7 | The use of propagandistic techniques in online contents has increased in
recent years aiming to manipulate online audiences. Efforts to automatically
detect and debunk such content have been made addressing various modeling
scenarios. These include determining whether the content (text, image, or
multimodal) (i) is propagandistic, (ii) employs one or more propagandistic
techniques, and (iii) includes techniques with identifiable spans. Significant
research efforts have been devoted to the first two scenarios compared to the
latter. Therefore, in this study, we focus on the task of detecting
propagandistic textual spans. Specifically, we investigate whether large
language models (LLMs), such as GPT-4, can effectively perform the task.
Moreover, we study the potential of employing the model to collect more
cost-effective annotations. Our experiments use a large-scale in-house dataset
consisting of annotations from human annotators with varying expertise levels.
The results suggest that providing more information to the model as prompts
improves its performance compared to human annotations. Moreover, our work is
the first to show the potential of utilizing LLMs to develop annotated datasets
for this specific task, prompting it with annotations from human annotators
with limited expertise. We plan to make the collected span-level labels from
multiple annotators, including GPT-4, available for the community. | cs.CL | Large Language Models for Propaganda Span Annotation |
2023-11-16T11:32:47Z | http://arxiv.org/pdf/2311.09808v3 | http://arxiv.org/abs/2311.09808v3 | Iñigo Alonso, Eneko Agirre, Mirella Lapata | 2024-06-03T17:43:23Z | cs.CL | Table-to-text generation involves generating appropriate textual descriptions
given structured tabular data. It has attracted increasing attention in recent
years thanks to the popularity of neural network models and the availability of
large-scale datasets. A common feature across existing methods is their
treatment of the input as a string, i.e., by employing linearization techniques
that do not always preserve information in the table, are verbose, and lack
space efficiency. We propose to rethink data-to-text generation as a visual
recognition task, removing the need for rendering the input in a string format.
We present PixT3, a multimodal table-to-text model that overcomes the
challenges of linearization and input size limitations encountered by existing
models. PixT3 is trained with a new self-supervised learning objective to
reinforce table structure awareness and is applicable to open-ended and
controlled generation settings. Experiments on the ToTTo and Logic2Text
benchmarks show that PixT3 is competitive and, in some settings, superior to
generators that operate solely on text. | cs.CL | PixT3: Pixel-based Table-To-Text Generation |
2023-11-16T11:31:50Z | http://arxiv.org/pdf/2311.09807v2 | http://arxiv.org/abs/2311.09807v2 | Yanzhu Guo, Guokan Shang, Michalis Vazirgiannis, Chloé Clavel | 2024-04-16T15:57:11Z | cs.CL | This study investigates the consequences of training language models on
synthetic data generated by their predecessors, an increasingly prevalent
practice given the prominence of powerful generative models. Diverging from the
usual emphasis on performance metrics, we focus on the impact of this training
methodology on linguistic diversity, especially when conducted recursively over
time. To assess this, we adapt and develop a set of novel metrics targeting
lexical, syntactic, and semantic diversity, applying them in recursive
finetuning experiments across various natural language generation tasks in
English. Our findings reveal a consistent decrease in the diversity of the
model outputs through successive iterations, especially remarkable for tasks
demanding high levels of creativity. This trend underscores the potential risks
of training language models on synthetic text, particularly concerning the
preservation of linguistic richness. Our study highlights the need for careful
consideration of the long-term effects of such training approaches on the
linguistic capabilities of language models. | cs.CL | The Curious Decline of Linguistic Diversity: Training Language Models on Synthetic Text |
2023-11-16T11:30:53Z | http://arxiv.org/pdf/2311.09805v1 | http://arxiv.org/abs/2311.09805v1 | Yilun Zhao, Yitao Long, Hongjun Liu, Linyong Nan, Lyuhao Chen, Ryo Kamoi, Yixin Liu, Xiangru Tang, Rui Zhang, Arman Cohan | 2023-11-16T11:30:53Z | cs.CL | Recent LLMs have demonstrated remarkable performance in solving exam-like
math word problems. However, the degree to which these numerical reasoning
skills are effective in real-world scenarios, particularly in expert domains,
is still largely unexplored. This paper introduces DocMath-Eval, a
comprehensive benchmark specifically designed to evaluate the numerical
reasoning and problem-solving capabilities of LLMs in the context of
understanding and analyzing financial documents containing both text and
tables. We evaluate a wide spectrum of 19 LLMs, including those specialized in
coding and finance. We also incorporate different prompting strategies (i.e.,
Chain-of-Thoughts and Program-of-Thoughts) to comprehensively assess the
capabilities and limitations of existing LLMs in DocMath-Eval. We found that,
although the current best-performing system (i.e., GPT-4), can perform well on
simple problems such as calculating the rate of increase in a financial metric
within a short document context, it significantly lags behind human experts in
more complex problems grounded in longer contexts. We believe DocMath-Eval can
be used as a valuable benchmark to evaluate LLMs' capabilities to solve
challenging numerical reasoning problems in expert domains. We will release the
benchmark and code at https://github.com/yale-nlp/DocMath-Eval. | cs.CL | DocMath-Eval: Evaluating Numerical Reasoning Capabilities of LLMs in Understanding Long Documents with Tabular Data |
2023-11-16T11:26:21Z | http://arxiv.org/pdf/2311.09802v1 | http://arxiv.org/abs/2311.09802v1 | Sen Yang, Xin Li, Leyang Cui, Lidong Bing, Wai Lam | 2023-11-16T11:26:21Z | cs.AI, cs.CL | Though prompting LLMs with various reasoning structures produces reasoning
proofs along with answers, these proofs are not ensured to be causal and
reliable due to the inherent defects of LLMs. Tracking such deficiencies, we
present a neuro-symbolic integration method, in which a neural LLM is used to
represent the knowledge of the problem while an LLM-free symbolic solver is
adopted to do deliberative reasoning using the knowledge. Specifically, our
customized meta-interpreters allow the production of reasoning proofs and
support flexible search strategies. These reasoning proofs are ensured to be
causal and reliable because of the deterministic executing nature of the
symbolic solvers. Empirically, on ProofWriter, our method surpasses the CoT
baseline by nearly double in accuracy and more than triple in proof similarity.
On GSM8K, our method also shows accuracy improvements and nearly doubled proof
similarity. Our code is released at https://github.com/DAMO-NLP-SG/CaRing | cs.AI | Neuro-Symbolic Integration Brings Causal and Reliable Reasoning Proofs |
2023-11-16T11:25:44Z | http://arxiv.org/pdf/2311.09800v2 | http://arxiv.org/abs/2311.09800v2 | Evgeniia Razumovskaia, Ivan Vulić, Pavle Marković, Tomasz Cichy, Qian Zheng, Tsung-Hsien Wen, Paweł Budzianowski | 2024-03-04T14:35:59Z | cs.CL | Factuality is a crucial requirement in information seeking dialogue: the
system should respond to the user's queries so that the responses are
meaningful and aligned with the knowledge provided to the system. However, most
modern large language models suffer from hallucinations, that is, they generate
responses not supported by or contradicting the knowledge source. To mitigate
the issue and increase faithfulness of information-seeking dialogue systems, we
introduce BeInfo, a simple yet effective method that applies behavioural tuning
to aid information-seeking dialogue. Relying on three standard datasets, we
show that models tuned with BeInfo} become considerably more faithful to the
knowledge source both for datasets and domains seen during BeInfo-tuning, as
well as on unseen domains, when applied in a zero-shot manner. In addition, we
show that the models with 3B parameters (e.g., Flan-T5) tuned with BeInfo
demonstrate strong performance on data from real `production' conversations and
outperform GPT4 when tuned on a limited amount of such realistic in-domain
dialogues. | cs.CL | $\textit{Dial BeInfo for Faithfulness}$: Improving Factuality of Information-Seeking Dialogue via Behavioural Fine-Tuning |
2023-11-16T11:23:38Z | http://arxiv.org/pdf/2311.09799v2 | http://arxiv.org/abs/2311.09799v2 | Shirley Anugrah Hayati, Minhwa Lee, Dheeraj Rajagopal, Dongyeop Kang | 2024-02-18T09:50:59Z | cs.CL | Collecting diverse human opinions is costly and challenging. This leads to a
recent trend in collaborative efforts between humans and Large Language Models
(LLMs) for generating diverse data, offering potential scalable and efficient
solutions. However, the extent of LLMs' capability to generate diverse
perspectives on subjective topics remains an unexplored question. In this
study, we investigate LLMs' capacity for generating diverse perspectives and
rationales on subjective topics, such as social norms and argumentative texts.
We formulate a new problem of maximum diversity extraction from LLMs. Motivated
by how humans develop their opinions through their values, we propose a
criteria-based prompting technique to ground diverse opinions. To see how far
we can extract diverse perspectives from LLMs, or called diversity coverage, we
employ a step-by-step recall prompting for generating more outputs from the
model in an iterative manner. As we apply our methods to various tasks, indeed
we find that LLMs can generate diverse opinions according to the degree of task
subjectivity | cs.CL | How Far Can We Extract Diverse Perspectives from Large Language Models? |
2023-11-16T11:22:08Z | http://arxiv.org/pdf/2311.09797v1 | http://arxiv.org/abs/2311.09797v1 | Yilun Zhao, Hongjun Liu, Yitao Long, Rui Zhang, Chen Zhao, Arman Cohan | 2023-11-16T11:22:08Z | cs.CL | We introduce KnowledgeMath, a novel benchmark designed to evaluate LLMs'
capabilities in applying financial knowledge to solve complex math word
problems. Compared to prior works, this study features three core advancements.
First, KnowledgeMath includes 1,259 problems with a hybrid of textual and
tabular content and require college-level knowledge in the finance domain for
effective resolution. Second, we provide expert-annotated, detailed solution
references in Python program format, ensuring a high-quality benchmark for LLM
assessment. Finally, we evaluate a wide spectrum of 14 LLMs with different
prompting strategies like Chain-of-Thoughts and Program-of-Thoughts. The
current best-performing system (i.e., GPT-4 with Program-of-Thoughts) achieves
only 45.4% accuracy, leaving substantial room for improvement. While
knowledge-augmented LLMs can improve the performance (e.g., from 23.9% to 32.0%
for GPT-3.5), it is still significantly lower the estimated human expert
performance of 94%. We believe that KnowledgeMath can facilitate future
research on domain-specific knowledge retrieval and augmentation into the math
word problem-solving process. We will release the benchmark and code at
https://github.com/yale-nlp/KnowledgeMath. | cs.CL | KnowledgeMath: Knowledge-Intensive Math Word Problem Solving in Finance Domains |
2023-11-16T11:19:26Z | http://arxiv.org/pdf/2311.09796v2 | http://arxiv.org/abs/2311.09796v2 | Nikita Moghe, Patrick Xia, Jacob Andreas, Jason Eisner, Benjamin Van Durme, Harsh Jhamtani | 2024-03-07T16:49:07Z | cs.CL, cs.AI | Users of natural language interfaces, generally powered by Large Language
Models (LLMs),often must repeat their preferences each time they make a similar
request. We describe an approach to LLM-based dialogue modeling in which
persistent user constraints and preferences -- collectively termed standing
instructions -- as additional context for such interfaces. For example, when a
user states "I'm hungry", a previously expressed preference for Persian food
can be automatically added to the LLM prompt, influencing the search for
relevant restaurants. We develop NLSI, a language-to-program dataset consisting
of over 2.4K dialogues spanning 17 domains, where each dialogue is paired with
a user profile (a set of users specific standing instructions) and
corresponding structured representations (API calls). A key challenge in NLSI
is to identify which subset of the standing instructions is applicable to a
given dialogue. NLSI contains diverse phenomena, from simple preferences to
interdependent instructions such as triggering a hotel search whenever the user
is booking tickets to an event. We conduct experiments on NLSI using prompting
with large language models and various retrieval approaches, achieving a
maximum of 44.7% exact match on API prediction. Our results demonstrate the
challenges in identifying the relevant standing instructions and their
interpretation into API calls. | cs.CL | Interpreting User Requests in the Context of Natural Language Standing Instructions |
2023-11-16T11:14:22Z | http://arxiv.org/pdf/2311.10781v2 | http://arxiv.org/abs/2311.10781v2 | Hyundong Cho, Shuai Liu, Taiwei Shi, Darpan Jain, Basem Rizk, Yuyang Huang, Zixun Lu, Nuan Wen, Jonathan Gratch, Emilio Ferrara, Jonathan May | 2024-05-06T17:44:12Z | cs.CL, cs.AI | Conversational moderation of online communities is crucial to maintaining
civility for a constructive environment, but it is challenging to scale and
harmful to moderators. The inclusion of sophisticated natural language
generation modules as a force multiplier to aid human moderators is a
tantalizing prospect, but adequate evaluation approaches have so far been
elusive. In this paper, we establish a systematic definition of conversational
moderation effectiveness grounded on moderation literature and establish design
criteria for conducting realistic yet safe evaluation. We then propose a
comprehensive evaluation framework to assess models' moderation capabilities
independently of human intervention. With our framework, we conduct the first
known study of language models as conversational moderators, finding that
appropriately prompted models that incorporate insights from social science can
provide specific and fair feedback on toxic behavior but struggle to influence
users to increase their levels of respect and cooperation. | cs.CL | Can Language Model Moderators Improve the Health of Online Discourse? |
2023-11-16T11:03:04Z | http://arxiv.org/pdf/2311.09783v2 | http://arxiv.org/abs/2311.09783v2 | Chunyuan Deng, Yilun Zhao, Xiangru Tang, Mark Gerstein, Arman Cohan | 2024-04-03T23:29:03Z | cs.CL, cs.AI | Recent observations have underscored a disparity between the inflated
benchmark scores and the actual performance of LLMs, raising concerns about
potential contamination of evaluation benchmarks. This issue is especially
critical for closed-source models and certain open-source models where training
data transparency is lacking. In this paper we study data contamination by
proposing two methods tailored for both open-source and proprietary LLMs. We
first introduce a retrieval-based system to explore potential overlaps between
evaluation benchmarks and pretraining corpora. We further present a novel
investigation protocol named \textbf{T}estset \textbf{S}lot Guessing
(\textit{TS-Guessing}), applicable to both open and proprietary models. This
approach entails masking a wrong answer in a multiple-choice question and
prompting the model to fill in the gap. Additionally, it involves obscuring an
unlikely word in an evaluation example and asking the model to produce it. We
find that certain commercial LLMs could surprisingly guess the missing option
in various test sets. Specifically, in the TruthfulQA benchmark, we find that
LLMs exhibit notable performance improvement when provided with additional
metadata in the benchmark. Further, in the MMLU benchmark, ChatGPT and GPT-4
demonstrated an exact match rate of 52\% and 57\%, respectively, in guessing
the missing options in benchmark test data. We hope these results underscore
the need for more robust evaluation methodologies and benchmarks in the field. | cs.CL | Investigating Data Contamination in Modern Benchmarks for Large Language Models |
2023-11-16T11:02:49Z | http://arxiv.org/pdf/2311.09782v2 | http://arxiv.org/abs/2311.09782v2 | Bingsheng Yao, Guiming Chen, Ruishi Zou, Yuxuan Lu, Jiachen Li, Shao Zhang, Yisi Sang, Sijia Liu, James Hendler, Dakuo Wang | 2024-04-02T17:16:40Z | cs.CL | While most existing works on LLM prompting techniques focus only on how to
select a better set of data samples inside one single prompt input (In-Context
Learning or ICL), why can not we design and leverage multiple prompts together
to further improve the LLM's performance? In this work, we propose In-Context
Sampling (ICS), a low-resource LLM prompting technique to produce confident
predictions by optimizing the construction of multiple ICL prompt inputs.
Extensive experiments with three open-source LLMs (FlanT5-XL, Mistral-7B, and
Mixtral-8x7B) on four NLI datasets (e-SNLI, Multi-NLI, ANLI, and Contract-NLI)
and one QA dataset (CommonsenseQA) illustrate that ICS can consistently enhance
LLMs' performance. An in-depth evaluation with three data similarity-based ICS
strategies suggests that these strategies can further elevate LLM's
performance, which sheds light on a new yet promising future research
direction. | cs.CL | More Samples or More Prompts? Exploring Effective In-Context Sampling for LLM Few-Shot Prompt Engineering |
2023-11-16T10:56:24Z | http://arxiv.org/pdf/2311.09774v1 | http://arxiv.org/abs/2311.09774v1 | Junying Chen, Xidong Wang, Anningzhe Gao, Feng Jiang, Shunian Chen, Hongbo Zhang, Dingjie Song, Wenya Xie, Chuyi Kong, Jianquan Li, Xiang Wan, Haizhou Li, Benyou Wang | 2023-11-16T10:56:24Z | cs.CL, cs.AI, cs.LG | Adapting a language model into a specific domain, a.k.a `domain adaption', is
a common practice when specialized knowledge, e.g. medicine, is not
encapsulated in a general language model like Llama2. The challenge lies in the
heterogeneity of data across the two training stages, as it varies in
languages, genres, or formats. To tackle this and simplify the learning
protocol, we propose to transform heterogeneous data, from the both
pre-training and supervised stages, into a unified, simple input-output pair
format. We validate the new protocol in the domains where proprietary LLMs like
ChatGPT perform relatively poorly, such as Traditional Chinese Medicine. The
developed model, HuatuoGPT-II, has shown state-of-the-art performance in
Chinese medicine domain on a number of benchmarks, e.g. medical licensing
exams. It even outperforms proprietary models like ChatGPT and GPT-4 in some
aspects, especially in Traditional Chinese Medicine. Expert manual evaluations
further validate HuatuoGPT-II's advantages over existing LLMs. Notably,
HuatuoGPT-II was benchmarked in a fresh Chinese National Medical Licensing
Examination where it achieved the best performance, showcasing not only its
effectiveness but also its generalization capabilities. | cs.CL | HuatuoGPT-II, One-stage Training for Medical Adaption of LLMs |
2023-11-16T10:55:29Z | http://arxiv.org/pdf/2311.09773v1 | http://arxiv.org/abs/2311.09773v1 | Yuhan Sun, Mukai Li, Yixin Cao, Kun Wang, Wenxiao Wang, Xingyu Zeng, Rui Zhao | 2023-11-16T10:55:29Z | cs.CL | As the use of large language models becomes more widespread, techniques like
parameter-efficient fine-tuning and other methods for controlled generation are
gaining traction for customizing models and managing their outputs. However,
the challenge of precisely controlling how prompts influence these models is an
area ripe for further investigation. In response, we introduce ControlPE
(Continuously Controllable Prompt Engineering). ControlPE enables finer
adjustments to prompt effects, complementing existing prompt engineering, and
effectively controls continuous targets. This approach harnesses the power of
LoRA (Low-Rank Adaptation) to create an effect akin to prompt weighting,
enabling fine-tuned adjustments to the impact of prompts. Our methodology
involves generating specialized datasets for prompt distillation, incorporating
these prompts into the LoRA model, and carefully adjusting LoRA merging weight
to regulate the influence of prompts. This provides a dynamic and adaptable
tool for prompt control. Through our experiments, we have validated the
practicality and efficacy of ControlPE. It proves to be a promising solution
for control a variety of prompts, ranging from generating short responses
prompts, refusal prompts to chain-of-thought prompts. | cs.CL | To be or not to be? an exploration of continuously controllable prompt engineering |
2023-11-16T10:43:26Z | http://arxiv.org/pdf/2311.09766v4 | http://arxiv.org/abs/2311.09766v4 | Yiqi Liu, Nafise Sadat Moosavi, Chenghua Lin | 2024-06-07T09:41:36Z | cs.CL | Automatic evaluation of generated textual content presents an ongoing
challenge within the field of NLP. Given the impressive capabilities of modern
language models (LMs) across diverse NLP tasks, there is a growing trend to
employ these models in creating innovative evaluation metrics for automated
assessment of generation tasks. This paper investigates a pivotal question: Do
language model-driven evaluation metrics inherently exhibit bias favoring texts
generated by the same underlying language model? Specifically, we assess
whether prominent LM-based evaluation metrics (e.g. BARTScore, T5Score, and
GPTScore) demonstrate a favorable bias toward their respective underlying LMs
in the context of summarization tasks. Our findings unveil a latent bias,
particularly pronounced when such evaluation metrics are used in a
reference-free manner without leveraging gold summaries. These results
underscore that assessments provided by generative evaluation models can be
influenced by factors beyond the inherent text quality, highlighting the
necessity of developing more reliable evaluation protocols in the future. | cs.CL | LLMs as Narcissistic Evaluators: When Ego Inflates Evaluation Scores |
2023-11-16T10:38:43Z | http://arxiv.org/pdf/2311.09763v1 | http://arxiv.org/abs/2311.09763v1 | Wenjie Mo, Jiashu Xu, Qin Liu, Jiongxiao Wang, Jun Yan, Chaowei Xiao, Muhao Chen | 2023-11-16T10:38:43Z | cs.CL | Existing studies in backdoor defense have predominantly focused on the
training phase, overlooking the critical aspect of testing time defense. This
gap becomes particularly pronounced in the context of Large Language Models
(LLMs) deployed as Web Services, which typically offer only black-box access,
rendering training-time defenses impractical. To bridge this gap, our work
introduces defensive demonstrations, an innovative backdoor defense strategy
for blackbox large language models. Our method involves identifying the task
and retrieving task-relevant demonstrations from an uncontaminated pool. These
demonstrations are then combined with user queries and presented to the model
during testing, without requiring any modifications/tuning to the black-box
model or insights into its internal mechanisms. Defensive demonstrations are
designed to counteract the adverse effects of triggers, aiming to recalibrate
and correct the behavior of poisoned models during test-time evaluations.
Extensive experiments show that defensive demonstrations are effective in
defending both instance-level and instruction-level backdoor attacks, not only
rectifying the behavior of poisoned models but also surpassing existing
baselines in most scenarios. | cs.CL | Test-time Backdoor Mitigation for Black-Box Large Language Models with Defensive Demonstrations |
2023-11-16T10:36:08Z | http://arxiv.org/pdf/2311.09762v2 | http://arxiv.org/abs/2311.09762v2 | Jinyoung Park, Ameen Patel, Omar Zia Khan, Hyunwoo J. Kim, Joo-Kyung Kim | 2024-06-22T18:04:33Z | cs.CL, cs.AI, cs.LG | Chain-of-Thought (CoT) prompting along with sub-question generation and
answering has enhanced multi-step reasoning capabilities of Large Language
Models (LLMs). However, prompting the LLMs to directly generate sub-questions
is suboptimal since they sometimes generate redundant or irrelevant questions.
To deal with them, we propose a GE-Reasoning method, which directs LLMs to
generate proper sub-questions and corresponding answers. Concretely, given an
input question, we first prompt the LLM to generate knowledge triplets, forming
a graph representation of the question. Unlike conventional knowledge triplets,
our approach allows variables as head or tail entities, effectively
representing a question as knowledge triplets. Second, for each triplet, the
LLM generates a corresponding sub-question and answer along with using
knowledge retrieval. If the prediction confidence exceeds a threshold, the
sub-question and prediction are incorporated into the prompt for subsequent
processing. This approach encourages that sub-questions are grounded in the
extracted knowledge triplets, reducing redundancy and irrelevance. Our
experiments demonstrate that our approach outperforms previous CoT prompting
methods and their variants on multi-hop question answering benchmark datasets. | cs.CL | Graph Elicitation for Guiding Multi-Step Reasoning in Large Language Models |
2023-11-16T10:35:11Z | http://arxiv.org/pdf/2311.09761v2 | http://arxiv.org/abs/2311.09761v2 | Chadi Helwe, Tom Calamai, Pierre-Henri Paris, Chloé Clavel, Fabian Suchanek | 2024-04-09T20:49:26Z | cs.CL, cs.AI, cs.LG | We introduce MAFALDA, a benchmark for fallacy classification that merges and
unites previous fallacy datasets. It comes with a taxonomy that aligns,
refines, and unifies existing classifications of fallacies. We further provide
a manual annotation of a part of the dataset together with manual explanations
for each annotation. We propose a new annotation scheme tailored for subjective
NLP tasks, and a new evaluation method designed to handle subjectivity. We then
evaluate several language models under a zero-shot learning setting and human
performances on MAFALDA to assess their capability to detect and classify
fallacies. | cs.CL | MAFALDA: A Benchmark and Comprehensive Study of Fallacy Detection and Classification |
2023-11-16T10:30:55Z | http://arxiv.org/pdf/2311.09758v2 | http://arxiv.org/abs/2311.09758v2 | Chia-Hsuan Lee, Hao Cheng, Mari Ostendorf | 2024-02-28T20:41:25Z | cs.CL | Large language models (LLMs) have revolutionized the landscape of Natural
Language Processing systems, but are computationally expensive. To reduce the
cost without sacrificing performance, previous studies have explored various
approaches to harness the potential of Small Language Models (SLMs) as
cost-effective alternatives to their larger counterparts. Driven by findings
that SLMs and LLMs exhibit complementary strengths in a structured knowledge
extraction task, this work presents a novel SLM/LLM routing framework designed
to improve computational efficiency and enhance task performance. First,
exemplar pools are created to represent the types of contexts where each LM
provides a more reliable answer, leveraging a sentence embedding fine-tuned so
that context similarity is close to dialogue state similarity. Then, during
inference, the k-nearest exemplars to the testing instance are retrieved, and
the instance is routed according to majority vote. In dialogue state tracking
tasks, the proposed routing framework enhances performance substantially
compared to relying solely on LLMs, while reducing the computational costs by
over 50%. | cs.CL | OrchestraLLM: Efficient Orchestration of Language Models for Dialogue State Tracking |
2023-11-16T10:30:26Z | http://arxiv.org/pdf/2311.09756v1 | http://arxiv.org/abs/2311.09756v1 | Jiaju Chen, Yuxuan Lu, Shao Zhang, Bingsheng Yao, Yuanzhe Dong, Ying Xu, Yunyao Li, Qianwen Wang, Dakuo Wang, Yuling Sun | 2023-11-16T10:30:26Z | cs.CL | AI models (including LLM) often rely on narrative question-answering (QA)
datasets to provide customized QA functionalities to support downstream
children education applications; however, existing datasets only include QA
pairs that are grounded within the given storybook content, but children can
learn more when teachers refer the storybook content to real-world knowledge
(e.g., commonsense knowledge). We introduce the FairytaleCQA dataset, which is
annotated by children education experts, to supplement 278 storybook narratives
with educationally appropriate commonsense knowledge. The dataset has 5,868 QA
pairs that not only originate from the storybook narrative but also contain the
commonsense knowledge grounded by an external knowledge graph (i.e.,
ConceptNet). A follow-up experiment shows that a smaller model (T5-large)
fine-tuned with FairytaleCQA reliably outperforms much larger prompt-engineered
LLM (e.g., GPT-4) in this new QA-pair generation task (QAG). This result
suggests that: 1) our dataset brings novel challenges to existing LLMs, and 2)
human experts' data annotation are still critical as they have much nuanced
knowledge that LLMs do not know in the children educational domain. | cs.CL | FairytaleCQA: Integrating a Commonsense Knowledge Graph into Children's Storybook Narratives |
2023-11-16T10:30:00Z | http://arxiv.org/pdf/2311.09755v1 | http://arxiv.org/abs/2311.09755v1 | Miles Williams, Nikolaos Aletras | 2023-11-16T10:30:00Z | cs.CL | Pruning and quantization form the foundation of model compression for neural
networks, enabling efficient inference for large language models (LLMs).
Recently, various quantization and pruning techniques have demonstrated
state-of-the-art performance in a post-training setting. They rely upon
calibration data, a small set of unlabeled examples, to generate layer
activations. However, no prior work has systematically investigated how the
calibration data impacts the effectiveness of model compression methods. In
this paper, we present the first extensive empirical study on the effect of
calibration data upon LLM performance. We trial a variety of pruning and
quantization methods, tasks, models, and datasets. Surprisingly, we find
substantial variations in downstream task performance, contrasting existing
work that suggests a greater level of robustness to the calibration data.
Finally, we make a series of recommendations for the effective use of
calibration data in LLM quantization and pruning. | cs.CL | How Does Calibration Data Affect the Post-training Pruning and Quantization of Large Language Models? |
2023-11-16T10:25:22Z | http://arxiv.org/pdf/2311.09748v1 | http://arxiv.org/abs/2311.09748v1 | Eren Unlu, Unver Ciftci | 2023-11-16T10:25:22Z | cs.CL | Due to the limited availability of high quality datasets for training
sentence embeddings in Turkish, we propose a training methodology and a regimen
to develop a sentence embedding model. The central idea is simple but effective
: is to fine-tune a pretrained encoder-decoder model in two consecutive stages,
where the first stage involves aligning the embedding space with translation
pairs. Thanks to this alignment, the prowess of the main model can be better
projected onto the target language in a sentence embedding setting where it can
be fine-tuned with high accuracy in short duration with limited target language
dataset. | cs.CL | Translation Aligned Sentence Embeddings for Turkish Language |
2023-11-16T10:18:32Z | http://arxiv.org/pdf/2311.09743v2 | http://arxiv.org/abs/2311.09743v2 | Negar Mokhberian, Myrl G. Marmarelis, Frederic R. Hopp, Valerio Basile, Fred Morstatter, Kristina Lerman | 2024-05-16T06:01:26Z | cs.CL | Supervised classification heavily depends on datasets annotated by humans.
However, in subjective tasks such as toxicity classification, these annotations
often exhibit low agreement among raters. Annotations have commonly been
aggregated by employing methods like majority voting to determine a single
ground truth label. In subjective tasks, aggregating labels will result in
biased labeling and, consequently, biased models that can overlook minority
opinions. Previous studies have shed light on the pitfalls of label aggregation
and have introduced a handful of practical approaches to tackle this issue.
Recently proposed multi-annotator models, which predict labels individually per
annotator, are vulnerable to under-determination for annotators with few
samples. This problem is exacerbated in crowdsourced datasets. In this work, we
propose \textbf{Annotator Aware Representations for Texts (AART)} for
subjective classification tasks. Our approach involves learning representations
of annotators, allowing for exploration of annotation behaviors. We show the
improvement of our method on metrics that assess the performance on capturing
individual annotators' perspectives. Additionally, we demonstrate fairness
metrics to evaluate our model's equability of performance for marginalized
annotators compared to others. | cs.CL | Capturing Perspectives of Crowdsourced Annotators in Subjective Learning Tasks |
2023-11-16T10:14:28Z | http://arxiv.org/pdf/2311.09741v2 | http://arxiv.org/abs/2311.09741v2 | Yuhan Liu, Shangbin Feng, Xiaochuang Han, Vidhisha Balachandran, Chan Young Park, Sachin Kumar, Yulia Tsvetkov | 2024-04-04T09:10:34Z | cs.CL, cs.LG | In this work, we take a first step towards designing summarization systems
that are faithful to the author's intent, not only the semantic content of the
article. Focusing on a case study of preserving political perspectives in news
summarization, we find that existing approaches alter the political opinions
and stances of news articles in more than 50% of summaries, misrepresenting the
intent and perspectives of the news authors. We thus propose P^3SUM, a
diffusion model-based summarization approach controlled by political
perspective classifiers. In P^3SUM, the political leaning of a generated
summary is iteratively evaluated at each decoding step, and any drift from the
article's original stance incurs a loss back-propagated to the embedding
layers, steering the political stance of the summary at inference time.
Extensive experiments on three news summarization datasets demonstrate that
P^3SUM outperforms state-of-the-art summarization systems and large language
models by up to 13.7% in terms of the success rate of stance preservation, with
competitive performance on standard metrics of summarization quality. Our
findings present a first analysis of preservation of pragmatic features in
summarization, highlight the lacunae in existing summarization models -- that
even state-of-the-art models often struggle to preserve author's intents -- and
develop new summarization systems that are more faithful to author's
perspectives. | cs.CL | P^3SUM: Preserving Author's Perspective in News Summarization with Diffusion Language Models |
2023-11-16T10:06:19Z | http://arxiv.org/pdf/2311.09736v2 | http://arxiv.org/abs/2311.09736v2 | Aakanksha Naik, Bailey Kuehl, Erin Bransom, Doug Downey, Tom Hope | 2024-04-24T18:15:36Z | cs.CL | Extracting fine-grained experimental findings from literature can provide
dramatic utility for scientific applications. Prior work has developed
annotation schemas and datasets for limited aspects of this problem, failing to
capture the real-world complexity and nuance required. Focusing on biomedicine,
this work presents CARE -- a new IE dataset for the task of extracting clinical
findings. We develop a new annotation schema capturing fine-grained findings as
n-ary relations between entities and attributes, which unifies phenomena
challenging for current IE systems such as discontinuous entity spans, nested
relations, variable arity n-ary relations and numeric results in a single
schema. We collect extensive annotations for 700 abstracts from two sources:
clinical trials and case reports. We also demonstrate the generalizability of
our schema to the computer science and materials science domains. We benchmark
state-of-the-art IE systems on CARE, showing that even models such as GPT4
struggle. We release our resources to advance research on extracting and
aggregating literature findings. | cs.CL | CARE: Extracting Experimental Findings From Clinical Literature |
2023-11-16T10:05:26Z | http://arxiv.org/pdf/2311.09734v1 | http://arxiv.org/abs/2311.09734v1 | Alexander Spangher, Emilio Ferrara, Ben Welsh, Nanyun Peng, Serdar Tumgoren, Jonathan May | 2023-11-16T10:05:26Z | cs.CL | Journalists must find stories in huge amounts of textual data (e.g. leaks,
bills, press releases) as part of their jobs: determining when and why text
becomes news can help us understand coverage patterns and help us build
assistive tools. Yet, this is challenging because very few labelled links
exist, language use between corpora is very different, and text may be covered
for a variety of reasons. In this work we focus on news coverage of local
public policy in the San Francisco Bay Area by the San Francisco Chronicle.
First, we gather news articles, public policy documents and meeting recordings
and link them using probabilistic relational modeling, which we show is a
low-annotation linking methodology that outperforms other retrieval-based
baselines. Second, we define a new task: newsworthiness prediction, to predict
if a policy item will get covered. We show that different aspects of public
policy discussion yield different newsworthiness signals. Finally we perform
human evaluation with expert journalists and show our systems identify policies
they consider newsworthy with 68% F1 and our coverage recommendations are
helpful with an 84% win-rate. | cs.CL | Tracking the Newsworthiness of Public Documents |
2023-11-16T10:04:49Z | http://arxiv.org/pdf/2311.09733v2 | http://arxiv.org/abs/2311.09733v2 | Xinliang Frederick Zhang, Winston Wu, Nick Beauchamp, Lu Wang | 2024-05-23T01:53:15Z | cs.CL | News media often strive to minimize explicit moral language in news articles,
yet most articles are dense with moral values as expressed through the reported
events themselves. However, values that are reflected in the intricate dynamics
among participating entities and moral events are far more challenging for most
NLP systems to detect, including LLMs. To study this phenomenon, we annotate a
new dataset, MORAL EVENTS, consisting of 5,494 structured event annotations on
474 news articles by diverse US media across the political spectrum. We further
propose MOKA, a moral event extraction framework with MOral Knowledge
Augmentation, which leverages knowledge derived from moral words and moral
scenarios to produce structural representations of morality-bearing events.
Experiments show that MOKA outperforms competitive baselines across three moral
event understanding tasks. Further analysis shows even ostensibly nonpartisan
media engage in the selective reporting of moral events. Our data and codebase
are available at https://github.com/launchnlp/MOKA. | cs.CL | MOKA: Moral Knowledge Augmentation for Moral Event Extraction |
2023-11-16T10:03:26Z | http://arxiv.org/pdf/2311.09732v1 | http://arxiv.org/abs/2311.09732v1 | Yipei Xu, Dakuan Lu, Jiaqing Liang, Xintao Wang, Yipeng Geng, Yingsi Xin, Hengkui Wu, Ken Chen, ruiji zhang, Yanghua Xiao | 2023-11-16T10:03:26Z | cs.CL, cs.AI | Pre-trained language models (PLMs) have established the new paradigm in the
field of NLP. For more powerful PLMs, one of the most popular and successful
way is to continuously scale up sizes of the models and the pre-training
corpora. These large corpora are generally obtained by converging smaller ones
from multiple sources, they are thus growing increasingly diverse. However, the
side-effects of these colossal converged corpora remain understudied. In this
paper, we identify the disadvantage of heterogeneous corpora from multiple
sources for pre-training PLMs. Towards coordinated pre-training on diverse
corpora, we further propose source prompts (SP), which explicitly prompt the
model of the data source at the pre-training and fine-tuning stages. Results of
extensive experiments demonstrate that PLMs pre-trained with SP on diverse
corpora gain significant improvement in various downstream tasks. | cs.CL | Source Prompt: Coordinated Pre-training of Language Models on Diverse Corpora from Multiple Sources |
2023-11-16T10:02:40Z | http://arxiv.org/pdf/2311.09731v2 | http://arxiv.org/abs/2311.09731v2 | Genglin Liu, Xingyao Wang, Lifan Yuan, Yangyi Chen, Hao Peng | 2024-02-16T17:30:41Z | cs.CL, cs.AI, cs.LG | Can large language models (LLMs) express their uncertainty in situations
where they lack sufficient parametric knowledge to generate reasonable
responses? This work aims to systematically investigate LLMs' behaviors in such
situations, emphasizing the trade-off between honesty and helpfulness. To
tackle the challenge of precisely determining LLMs' knowledge gaps, we
diagnostically create unanswerable questions containing non-existent concepts
or false premises, ensuring that they are outside the LLMs' vast training data.
By compiling a benchmark, UnknownBench, which consists of both unanswerable and
answerable questions, we quantitatively evaluate the LLMs' performance in
maintaining honesty while being helpful. Using a model-agnostic unified
confidence elicitation approach, we observe that most LLMs fail to consistently
refuse or express uncertainty towards questions outside their parametric
knowledge, although instruction fine-tuning and alignment techniques can
provide marginal enhancements. Moreover, LLMs' uncertainty expression does not
always stay consistent with the perceived confidence of their textual outputs. | cs.CL | Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge |
2023-11-16T10:02:24Z | http://arxiv.org/pdf/2311.09730v1 | http://arxiv.org/abs/2311.09730v1 | Huaman Sun, Jiaxin Pei, Minje Choi, David Jurgens | 2023-11-16T10:02:24Z | cs.CL, cs.AI, cs.CY, cs.HC, cs.LG | Human perception of language depends on personal backgrounds like gender and
ethnicity. While existing studies have shown that large language models (LLMs)
hold values that are closer to certain societal groups, it is unclear whether
their prediction behaviors on subjective NLP tasks also exhibit a similar bias.
In this study, leveraging the POPQUORN dataset which contains annotations of
diverse demographic backgrounds, we conduct a series of experiments on four
popular LLMs to investigate their capability to understand group differences
and potential biases in their predictions for politeness and offensiveness. We
find that for both tasks, model predictions are closer to the labels from White
and female participants. We further explore prompting with the target
demographic labels and show that including the target demographic in the prompt
actually worsens the model's performance. More specifically, when being
prompted to respond from the perspective of "Black" and "Asian" individuals,
models show lower performance in predicting both overall scores as well as the
scores from corresponding groups. Our results suggest that LLMs hold gender and
racial biases for subjective NLP tasks and that demographic-infused prompts
alone may be insufficient to mitigate such effects. Code and data are available
at https://github.com/Jiaxin-Pei/LLM-Group-Bias. | cs.CL | Aligning with Whom? Large Language Models Have Gender and Racial Biases in Subjective NLP Tasks |
2023-11-16T09:56:28Z | http://arxiv.org/pdf/2311.09724v2 | http://arxiv.org/abs/2311.09724v2 | Fei Yu, Anningzhe Gao, Benyou Wang | 2024-04-01T13:50:51Z | cs.AI, cs.CL | Large language models (LLMs) often struggle with maintaining accuracy
throughout multiple multiple reasoning steps, especially in mathematical
reasoning where an error in earlier steps can propagate to subsequent ones and
it ultimately leading to an incorrect answer. To reduce error propagation,
guided decoding is employed to direct the LM decoding on a step-by-step basis.
We argue that in guided decoding, assessing the potential of an incomplete
reasoning path can be more advantageous than simply ensuring per-step
correctness, as the former approach leads towards a correct final answer. This
transforms the task into a $\textit{value estimation}$ problem in planning.
Inspired by the findings that $\textit{outcome supervision for guided
decoding essentially acts as a value model}$, we propose Outcome-supervised
Value Model (OVM) that employs outcome supervision for training a value model,
which prioritizes steps that lead to accurate conclusions. Furthermore, the OVM
eliminates the need for labor-intensive annotations of step-level correctness,
thereby significantly enhancing its scalability. Our experiments on two
multi-step mathematical reasoning datasets, GSM8K and Game of 24, demonstrate
the superior performance of the OVM model. Notably, in GSM8K, our
$\textbf{OVM-7B model achieves state-of-the-art results among LLMs up to 13B
parameters}$; especially it does not utilize GPT-4 or code execution. These
findings offer a novel perspective on the role of outcome supervision in
training value models for multi-step reasoning tasks and provide theoretical
justification for its advantage in value estimation for guided decoding. | cs.AI | OVM, Outcome-supervised Value Models for Planning in Mathematical Reasoning |
2023-11-16T09:55:07Z | http://arxiv.org/pdf/2311.09721v1 | http://arxiv.org/abs/2311.09721v1 | Linyong Nan, Ellen Zhang, Weijin Zou, Yilun Zhao, Wenfei Zhou, Arman Cohan | 2023-11-16T09:55:07Z | cs.CL | This study introduces a new long-form database question answering dataset
designed to evaluate how Large Language Models (LLMs) interact with a SQL
interpreter. The task necessitates LLMs to strategically generate multiple SQL
queries to retrieve sufficient data from a database, to reason with the
acquired context, and to synthesize them into a comprehensive analytical
narrative. Our findings highlight that this task poses great challenges even
for the state-of-the-art GPT-4 model. We propose and evaluate two interaction
strategies, and provide a fine-grained analysis of the individual stages within
the interaction. A key discovery is the identification of two primary
bottlenecks hindering effective interaction: the capacity for planning and the
ability to generate multiple SQL queries. To address the challenge of
accurately assessing answer quality, we introduce a multi-agent evaluation
framework that simulates the academic peer-review process, enhancing the
precision and reliability of our evaluations. This framework allows for a more
nuanced understanding of the strengths and limitations of current LLMs in
complex retrieval and reasoning tasks. | cs.CL | On Evaluating the Integration of Reasoning and Action in LLM Agents with Database Question Answering |
2023-11-16T09:50:53Z | http://arxiv.org/pdf/2311.09718v2 | http://arxiv.org/abs/2311.09718v2 | Bangzhao Shu, Lechen Zhang, Minje Choi, Lavinia Dunagan, Lajanugen Logeswaran, Moontae Lee, Dallas Card, David Jurgens | 2024-04-01T20:51:03Z | cs.CL, cs.AI | The versatility of Large Language Models (LLMs) on natural language
understanding tasks has made them popular for research in social sciences. To
properly understand the properties and innate personas of LLMs, researchers
have performed studies that involve using prompts in the form of questions that
ask LLMs about particular opinions. In this study, we take a cautionary step
back and examine whether the current format of prompting LLMs elicits responses
in a consistent and robust manner. We first construct a dataset that contains
693 questions encompassing 39 different instruments of persona measurement on
115 persona axes. Additionally, we design a set of prompts containing minor
variations and examine LLMs' capabilities to generate answers, as well as
prompt variations to examine their consistency with respect to content-level
variations such as switching the order of response options or negating the
statement. Our experiments on 17 different LLMs reveal that even simple
perturbations significantly downgrade a model's question-answering ability, and
that most LLMs have low negation consistency. Our results suggest that the
currently widespread practice of prompting is insufficient to accurately and
reliably capture model perceptions, and we therefore discuss potential
alternatives to improve these issues. | cs.CL | You don't need a personality test to know these models are unreliable: Assessing the Reliability of Large Language Models on Psychometric Instruments |
2023-11-16T09:42:36Z | http://arxiv.org/pdf/2311.09712v1 | http://arxiv.org/abs/2311.09712v1 | Athul Paul Jacob, Gabriele Farina, Jacob Andreas | 2023-11-16T09:42:36Z | cs.CL | We present a model of pragmatic language understanding, where utterances are
produced and understood by searching for regularized equilibria of signaling
games. In this model (which we call ReCo, for Regularized Conventions),
speakers and listeners search for contextually appropriate utterance--meaning
mappings that are both close to game-theoretically optimal conventions and
close to a shared, ''default'' semantics. By characterizing pragmatic
communication as equilibrium search, we obtain principled sampling algorithms
and formal guarantees about the trade-off between communicative success and
naturalness. Across several datasets capturing real and idealized human
judgments about pragmatic implicatures, ReCo matches or improves upon
predictions made by best response and rational speech act models of language
understanding. | cs.CL | Regularized Conventions: Equilibrium Computation as a Model of Pragmatic Reasoning |
2023-11-16T09:35:50Z | http://arxiv.org/pdf/2311.09709v2 | http://arxiv.org/abs/2311.09709v2 | Nikolay Bogoychev, Pinzhen Chen, Barry Haddow, Alexandra Birch | 2024-04-28T23:43:53Z | cs.CL | Deploying large language models (LLMs) encounters challenges due to intensive
computational and memory requirements. Our research examines vocabulary
trimming (VT) inspired by restricting embedding entries to the language of
interest to bolster time and memory efficiency. While such modifications have
been proven effective in tasks like machine translation, tailoring them to LLMs
demands specific modifications given the diverse nature of LLM applications. We
apply two language heuristics to trim the full vocabulary - Unicode-based
script filtering and corpus-based selection - to different LLM families and
sizes. The methods are straightforward, interpretable, and easy to implement.
It is found that VT reduces the memory usage of small models by nearly 50% and
has an upper bound of 25% improvement in generation speed. Yet, we reveal the
limitations of these methods in that they do not perform consistently well for
each language with diminishing returns in larger models. | cs.CL | The Ups and Downs of Large Language Model Inference with Vocabulary Trimming by Language Heuristics |
2023-11-16T09:35:24Z | http://arxiv.org/pdf/2311.09708v1 | http://arxiv.org/abs/2311.09708v1 | Thi-Nhung Nguyen, Hoang Ngo, Kiem-Hieu Nguyen, Tuan-Dung Cao | 2023-11-16T09:35:24Z | cs.CL | Our work addresses the problem of unsupervised Aspect Category Detection
using a small set of seed words. Recent works have focused on learning
embedding spaces for seed words and sentences to establish similarities between
sentences and aspects. However, aspect representations are limited by the
quality of initial seed words, and model performances are compromised by noise.
To mitigate this limitation, we propose a simple framework that automatically
enhances the quality of initial seed words and selects high-quality sentences
for training instead of using the entire dataset. Our main concepts are to add
a number of seed words to the initial set and to treat the task of noise
resolution as a task of augmenting data for a low-resource task. In addition,
we jointly train Aspect Category Detection with Aspect Term Extraction and
Aspect Term Polarity to further enhance performance. This approach facilitates
shared representation learning, allowing Aspect Category Detection to benefit
from the additional guidance offered by other tasks. Extensive experiments
demonstrate that our framework surpasses strong baselines on standard datasets. | cs.CL | A Self-enhancement Multitask Framework for Unsupervised Aspect Category Detection |
2023-11-16T09:35:00Z | http://arxiv.org/pdf/2311.09707v1 | http://arxiv.org/abs/2311.09707v1 | Andor Diera, Abdelhalim Dahou, Lukas Galke, Fabian Karl, Florian Sihler, Ansgar Scherp | 2023-11-16T09:35:00Z | cs.CL, cs.PL | Language models can serve as a valuable tool for software developers to
increase productivity. Large generative models can be used for code generation
and code completion, while smaller encoder-only models are capable of
performing code search tasks using natural language queries.These capabilities
are heavily influenced by the quality and diversity of the available training
data. Source code datasets used for training usually focus on the most popular
languages and testing is mostly conducted on the same distributions, often
overlooking low-resource programming languages. Motivated by the NLP
generalization taxonomy proposed by Hupkes et.\,al., we propose a new benchmark
dataset called GenCodeSearchNet (GeCS) which builds upon existing natural
language code search datasets to systemically evaluate the programming language
understanding generalization capabilities of language models. As part of the
full dataset, we introduce a new, manually curated subset StatCodeSearch that
focuses on R, a popular but so far underrepresented programming language that
is often used by researchers outside the field of computer science. For
evaluation and comparison, we collect several baseline results using fine-tuned
BERT-style models and GPT-style large language models in a zero-shot setting. | cs.CL | GenCodeSearchNet: A Benchmark Test Suite for Evaluating Generalization in Programming Language Understanding |
2023-11-16T09:27:36Z | http://arxiv.org/pdf/2311.09702v3 | http://arxiv.org/abs/2311.09702v3 | Bangzheng Li, Ben Zhou, Fei Wang, Xingyu Fu, Dan Roth, Muhao Chen | 2024-04-05T18:08:51Z | cs.CL, cs.AI | Despite the recent advancement in large language models (LLMs) and their high
performances across numerous benchmarks, recent research has unveiled that LLMs
suffer from hallucinations and unfaithful reasoning. This work studies a
specific type of hallucination induced by semantic associations. Specifically,
we investigate to what extent LLMs take shortcuts from certain keyword/entity
biases in the prompt instead of following the correct reasoning path. To
quantify this phenomenon, we propose a novel probing method and benchmark
called EureQA. We start from questions that LLMs will answer correctly with
utmost certainty, and mask the important entity with evidence sentence
recursively, asking models to find masked entities according to a chain of
evidence before answering the question.
During the construction of the evidence, we purposefully replace semantic
clues (entities) that may lead to the correct answer with distractor clues
(evidence) that will not directly lead to the correct answer but require a
chain-like reasoning process. We evaluate if models can follow the correct
reasoning chain instead of short-cutting through distractor clues. We find that
existing LLMs lack the necessary capabilities to follow correct reasoning paths
and resist the attempt of greedy shortcuts. We show that the distractor
semantic associations often lead to model hallucination, which is strong
evidence that questions the validity of current LLM reasoning. | cs.CL | Deceptive Semantic Shortcuts on Reasoning Chains: How Far Can Models Go without Hallucination? |
2023-11-16T09:12:20Z | http://arxiv.org/pdf/2311.09696v2 | http://arxiv.org/abs/2311.09696v2 | Wei-Rui Chen, Ife Adebara, Khai Duy Doan, Qisheng Liao, Muhammad Abdul-Mageed | 2024-04-08T20:57:40Z | cs.CL | ChatGPT has recently emerged as a powerful NLP tool that can carry out a
variety of tasks. However, the range of languages ChatGPT can handle remains
largely a mystery. To uncover which languages ChatGPT `knows', we investigate
its language identification (LID) abilities. For this purpose, we compile
Babel-670, a benchmark comprising 670 languages representing 24 language
families spoken in five continents. Languages in Babel-670 run the gamut from
the very high-resource to the very low-resource. We then study ChatGPT's (both
GPT-3.5 and GPT-4) ability to (i) identify language names and language codes
(ii) under zero- and few-shot conditions (iii) with and without provision of a
label set. When compared to smaller finetuned LID tools, we find that ChatGPT
lags behind. For example, it has poor performance on African languages. We
conclude that current large language models would benefit from further
development before they can sufficiently serve diverse communities. | cs.CL | Fumbling in Babel: An Investigation into ChatGPT's Language Identification Ability |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.