publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2024-03-20T09:21:32Z | http://arxiv.org/pdf/2403.13433v2 | http://arxiv.org/abs/2403.13433v2 | Zhouhong Gu, Xiaoxuan Zhu, Haoran Guo, Lin Zhang, Yin Cai, Hao Shen, Jiangjie Chen, Zheyu Ye, Yifei Dai, Yan Gao, Yao Hu, Hongwei Feng, Yanghua Xiao | 2024-04-04T07:40:31Z | cs.AI, cs.CL, cs.CY | Language significantly influences the formation and evolution of Human
emergent behavior, which is crucial in understanding collective intelligence
within human societies. Considering that the study of how language affects
human behavior needs to put it into the dynamic scenarios in which it is used,
we introduce AgentGroupChat in this paper, a simulation that delves into the
complex role of language in shaping collective behavior through interactive
debate scenarios. Central to this simulation are characters engaging in dynamic
conversation interactions. To enable simulation, we introduce the Verbal
Strategist Agent, utilizing large language models to enhance interaction
strategies by incorporating elements of persona and action. We set four
narrative scenarios based on AgentGroupChat to demonstrate the simulation's
capacity to mimic complex language use in group dynamics. Evaluations focus on
aligning agent behaviors with human expectations and the emergence of
collective behaviors within the simulation. Results reveal that emergent
behaviors materialize from a confluence of factors: a conducive environment for
extensive information exchange, characters with diverse traits, high linguistic
comprehension, and strategic adaptability. During discussions on ``the impact
of AI on humanity'' in AgentGroupChat simulation, philosophers commonly agreed
that ``AI could enhance societal welfare with judicious limitations'' and even
come to a conclusion that ``the essence of true intelligence encompasses
understanding the necessity to constrain self abilities''. Additionally, in the
competitive domain of casting for primary roles in films in AgentGroupChat,
certain actors were ready to reduce their remuneration or accept lesser roles,
motivated by their deep-seated desire to contribute to the project. | cs.AI | AgentGroupChat: An Interactive Group Chat Simulacra For Better Eliciting Emergent Behavior |
2024-03-20T08:52:40Z | http://arxiv.org/pdf/2403.15469v1 | http://arxiv.org/abs/2403.15469v1 | Shivam Ratnakant Mhaskar, Nirmesh J. Shah, Mohammadi Zaki, Ashishkumar P. Gudmalwar, Pankaj Wasnik, Rajiv Ratn Shah | 2024-03-20T08:52:40Z | cs.CL, cs.LG, eess.AS | Traditional Automatic Video Dubbing (AVD) pipeline consists of three key
modules, namely, Automatic Speech Recognition (ASR), Neural Machine Translation
(NMT), and Text-to-Speech (TTS). Within AVD pipelines, isometric-NMT algorithms
are employed to regulate the length of the synthesized output text. This is
done to guarantee synchronization with respect to the alignment of video and
audio subsequent to the dubbing process. Previous approaches have focused on
aligning the number of characters and words in the source and target language
texts of Machine Translation models. However, our approach aims to align the
number of phonemes instead, as they are closely associated with speech
duration. In this paper, we present the development of an isometric NMT system
using Reinforcement Learning (RL), with a focus on optimizing the alignment of
phoneme counts in the source and target language sentence pairs. To evaluate
our models, we propose the Phoneme Count Compliance (PCC) score, which is a
measure of length compliance. Our approach demonstrates a substantial
improvement of approximately 36% in the PCC score compared to the
state-of-the-art models when applied to English-Hindi language pairs. Moreover,
we propose a student-teacher architecture within the framework of our RL
approach to maintain a trade-off between the phoneme count and translation
quality. | cs.CL | Isometric Neural Machine Translation using Phoneme Count Ratio Reward-based Reinforcement Learning |
2024-03-20T08:38:25Z | http://arxiv.org/pdf/2405.02305v1 | http://arxiv.org/abs/2405.02305v1 | Yannis Tevissen, Khalil Guetari, Marine Tassel, Erwan Kerleroux, Frédéric Petitpont | 2024-03-20T08:38:25Z | cs.CV, cs.AI, cs.CL, eess.IV | Image captioning models are widely used to describe recent and archived
pictures with the objective of improving their accessibility and retrieval.
Yet, these approaches tend to be inefficient and biased at retrieving people's
names. In this work we introduce AstroCaptions, a dataset for the image
captioning task. This dataset specifically contains thousands of public
fig-ures that are complex to identify for a traditional model. We also propose
a novel post-processing method to insert identified people's names inside the
caption using explainable AI tools and the grounding capabilities of
vi-sion-language models. The results obtained with this method show
signifi-cant improvements of captions quality and a potential of reducing
halluci-nations. Up to 93.2% of the persons detected can be inserted in the
image captions leading to improvements in the BLEU, ROUGE, CIDEr and METEOR
scores of each captioning model. | cs.CV | Inserting Faces inside Captions: Image Captioning with Attention Guided Merging |
2024-03-20T08:08:54Z | http://arxiv.org/pdf/2403.13372v4 | http://arxiv.org/abs/2403.13372v4 | Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, Zheyan Luo, Zhangchi Feng, Yongqiang Ma | 2024-06-27T22:44:48Z | cs.CL, cs.AI | Efficient fine-tuning is vital for adapting large language models (LLMs) to
downstream tasks. However, it requires non-trivial efforts to implement these
methods on different models. We present LlamaFactory, a unified framework that
integrates a suite of cutting-edge efficient training methods. It provides a
solution for flexibly customizing the fine-tuning of 100+ LLMs without the need
for coding through the built-in web UI LlamaBoard. We empirically validate the
efficiency and effectiveness of our framework on language modeling and text
generation tasks. It has been released at
https://github.com/hiyouga/LLaMA-Factory and received over 25,000 stars and
3,000 forks. | cs.CL | LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models |
2024-03-20T08:01:33Z | http://arxiv.org/pdf/2403.13369v1 | http://arxiv.org/abs/2403.13369v1 | Phillip Richter-Pechanski, Philipp Wiesenbach, Dominic M. Schwab, Christina Kiriakou, Nicolas Geis, Christoph Dieterich, Anette Frank | 2024-03-20T08:01:33Z | cs.CL, cs.AI, cs.LG | Automatic extraction of medical information from clinical documents poses
several challenges: high costs of required clinical expertise, limited
interpretability of model predictions, restricted computational resources and
privacy regulations. Recent advances in domain-adaptation and prompting methods
showed promising results with minimal training data using lightweight masked
language models, which are suited for well-established interpretability
methods. We are first to present a systematic evaluation of these methods in a
low-resource setting, by performing multi-class section classification on
German doctor's letters. We conduct extensive class-wise evaluations supported
by Shapley values, to validate the quality of our small training data set and
to ensure the interpretability of model predictions. We demonstrate that a
lightweight, domain-adapted pretrained model, prompted with just 20 shots,
outperforms a traditional classification model by 30.5% accuracy. Our results
serve as a process-oriented guideline for clinical information extraction
projects working with low-resource. | cs.CL | Clinical information extraction for Low-resource languages with Few-shot learning using Pre-trained language models and Prompting |
2024-03-20T08:01:22Z | http://arxiv.org/pdf/2403.13368v1 | http://arxiv.org/abs/2403.13368v1 | Shaonan Wang, Jingyuan Sun, Yunhao Zhang, Nan Lin, Marie-Francine Moens, Chengqing Zong | 2024-03-20T08:01:22Z | cs.CL, cs.AI | Despite differing from the human language processing mechanism in
implementation and algorithms, current language models demonstrate remarkable
human-like or surpassing language capabilities. Should computational language
models be employed in studying the brain, and if so, when and how? To delve
into this topic, this paper reviews efforts in using computational models for
brain research, highlighting emerging trends. To ensure a fair comparison, the
paper evaluates various computational models using consistent metrics on the
same dataset. Our analysis reveals that no single model outperforms others on
all datasets, underscoring the need for rich testing datasets and rigid
experimental control to draw robust conclusions in studies involving
computational models. | cs.CL | Computational Models to Study Language Processing in the Human Brain: A Survey |
2024-03-20T07:44:06Z | http://arxiv.org/pdf/2403.13362v3 | http://arxiv.org/abs/2403.13362v3 | Hadi Askari, Anshuman Chhabra, Bernhard Clemm von Hohenberg, Michael Heseltine, Magdalena Wojcieszak | 2024-03-30T03:10:48Z | cs.SI, cs.AI, cs.CL | Polarization, declining trust, and wavering support for democratic norms are
pressing threats to U.S. democracy. Exposure to verified and quality news may
lower individual susceptibility to these threats and make citizens more
resilient to misinformation, populism, and hyperpartisan rhetoric. This project
examines how to enhance users' exposure to and engagement with verified and
ideologically balanced news in an ecologically valid setting. We rely on a
large-scale two-week long field experiment (from 1/19/2023 to 2/3/2023) on
28,457 Twitter users. We created 28 bots utilizing GPT-2 that replied to users
tweeting about sports, entertainment, or lifestyle with a contextual reply
containing two hardcoded elements: a URL to the topic-relevant section of
quality news organization and an encouragement to follow its Twitter account.
To further test differential effects by gender of the bots, treated users were
randomly assigned to receive responses by bots presented as female or male. We
examine whether our over-time intervention enhances the following of news media
organization, the sharing and the liking of news content and the tweeting about
politics and the liking of political content. We find that the treated users
followed more news accounts and the users in the female bot treatment were more
likely to like news content than the control. Most of these results, however,
were small in magnitude and confined to the already politically interested
Twitter users, as indicated by their pre-treatment tweeting about politics.
These findings have implications for social media and news organizations, and
also offer direction for future work on how Large Language Models and other
computational interventions can effectively enhance individual on-platform
engagement with quality news and public affairs. | cs.SI | Incentivizing News Consumption on Social Media Platforms Using Large Language Models and Realistic Bot Accounts |
2024-03-20T07:05:19Z | http://arxiv.org/pdf/2403.13344v1 | http://arxiv.org/abs/2403.13344v1 | Zhihan Zhou, Qixiang Fang, Leonardo Neves, Francesco Barbieri, Yozen Liu, Han Liu, Maarten W. Bos, Ron Dotsch | 2024-03-20T07:05:19Z | cs.SI, cs.AI, cs.CL, cs.HC, cs.IR, cs.LG | User embeddings play a crucial role in user engagement forecasting and
personalized services. Recent advances in sequence modeling have sparked
interest in learning user embeddings from behavioral data. Yet behavior-based
user embedding learning faces the unique challenge of dynamic user modeling. As
users continuously interact with the apps, user embeddings should be
periodically updated to account for users' recent and long-term behavior
patterns. Existing methods highly rely on stateless sequence models that lack
memory of historical behavior. They have to either discard historical data and
use only the most recent data or reprocess the old and new data jointly. Both
cases incur substantial computational overhead. To address this limitation, we
introduce User Stateful Embedding (USE). USE generates user embeddings and
reflects users' evolving behaviors without the need for exhaustive reprocessing
by storing previous model states and revisiting them in the future.
Furthermore, we introduce a novel training objective named future W-behavior
prediction to transcend the limitations of next-token prediction by forecasting
a broader horizon of upcoming user behaviors. By combining it with the Same
User Prediction, a contrastive learning-based objective that predicts whether
different segments of behavior sequences belong to the same user, we further
improve the embeddings' distinctiveness and representativeness. We conducted
experiments on 8 downstream tasks using Snapchat users' behavioral logs in both
static (i.e., fixed user behavior sequences) and dynamic (i.e., periodically
updated user behavior sequences) settings. We demonstrate USE's superior
performance over established baselines. The results underscore USE's
effectiveness and efficiency in integrating historical and recent user behavior
sequences into user embeddings in dynamic user modeling. | cs.SI | USE: Dynamic User Modeling with Stateful Sequence Models |
2024-03-20T06:37:59Z | http://arxiv.org/pdf/2403.13334v2 | http://arxiv.org/abs/2403.13334v2 | Chih-Wei Song, Yin-Te Tsai | 2024-03-26T12:24:46Z | cs.CL, cs.AI | This research's primary motivation of this study is to address the high
hardware and computational demands typically associated with LLMs.Therefore,our
goal is to find a balance between model lightness and performance,striving to
maximize performance while using a comparatively lightweight model. Hyacinth6B
was developed with this objective in mind,aiming to fully leverage the core
capabilities of LLMs without incurring substantial resource costs, effectively
pushing the boundaries of smaller model's performance. The training approach
involves parameter efficient finetuning using the LoRA method. | cs.CL | Hyacinth6B: A large language model for Traditional Chinese |
2024-03-20T06:28:09Z | http://arxiv.org/pdf/2403.15467v1 | http://arxiv.org/abs/2403.15467v1 | Seunguk Yu, Juhwan Choi, Youngbin Kim | 2024-03-20T06:28:09Z | cs.CL | Offensive language detection is an important task for filtering out abusive
expressions and improving online user experiences. However, malicious users
often attempt to avoid filtering systems through the involvement of textual
noises. In this paper, we propose these evasions as user-intended adversarial
attacks that insert special symbols or leverage the distinctive features of the
Korean language. Furthermore, we introduce simple yet effective pooling
strategies in a layer-wise manner to defend against the proposed attacks,
focusing on the preceding layers not just the last layer to capture both
offensiveness and token embeddings. We demonstrate that these pooling
strategies are more robust to performance degradation even when the attack rate
is increased, without directly training of such patterns. Notably, we found
that models pre-trained on clean texts could achieve a comparable performance
in detecting attacked offensive language, to models pre-trained on noisy texts
by employing these pooling strategies. | cs.CL | Don't be a Fool: Pooling Strategies in Offensive Language Detection from User-Intended Adversarial Attacks |
2024-03-20T05:34:03Z | http://arxiv.org/pdf/2403.13313v1 | http://arxiv.org/abs/2403.13313v1 | Subhabrata Mukherjee, Paul Gamble, Markel Sanz Ausin, Neel Kant, Kriti Aggarwal, Neha Manjunath, Debajyoti Datta, Zhengliang Liu, Jiayuan Ding, Sophia Busacca, Cezanne Bianco, Swapnil Sharma, Rae Lasko, Michelle Voisard, Sanchay Harneja, Darya Filippova, Gerry Meixiong, Kevin Cha, Amir Youssefi, Meyhaa Buvanesh, Howard Weingram, Sebastian Bierman-Lytle, Harpreet Singh Mangat, Kim Parikh, Saad Godil, Alex Miller | 2024-03-20T05:34:03Z | cs.AI, cs.CL | We develop Polaris, the first safety-focused LLM constellation for real-time
patient-AI healthcare conversations. Unlike prior LLM works in healthcare
focusing on tasks like question answering, our work specifically focuses on
long multi-turn voice conversations. Our one-trillion parameter constellation
system is composed of several multibillion parameter LLMs as co-operative
agents: a stateful primary agent that focuses on driving an engaging
conversation and several specialist support agents focused on healthcare tasks
performed by nurses to increase safety and reduce hallucinations. We develop a
sophisticated training protocol for iterative co-training of the agents that
optimize for diverse objectives. We train our models on proprietary data,
clinical care plans, healthcare regulatory documents, medical manuals, and
other medical reasoning documents. We align our models to speak like medical
professionals, using organic healthcare conversations and simulated ones
between patient actors and experienced nurses. This allows our system to
express unique capabilities such as rapport building, trust building, empathy
and bedside manner. Finally, we present the first comprehensive clinician
evaluation of an LLM system for healthcare. We recruited over 1100 U.S.
licensed nurses and over 130 U.S. licensed physicians to perform end-to-end
conversational evaluations of our system by posing as patients and rating the
system on several measures. We demonstrate Polaris performs on par with human
nurses on aggregate across dimensions such as medical safety, clinical
readiness, conversational quality, and bedside manner. Additionally, we conduct
a challenging task-based evaluation of the individual specialist support
agents, where we demonstrate our LLM agents significantly outperform a much
larger general-purpose LLM (GPT-4) as well as from its own medium-size class
(LLaMA-2 70B). | cs.AI | Polaris: A Safety-focused LLM Constellation Architecture for Healthcare |
2024-03-20T05:29:06Z | http://arxiv.org/pdf/2403.13312v1 | http://arxiv.org/abs/2403.13312v1 | Dongwei Jiang, Marcio Fonseca, Shay B. Cohen | 2024-03-20T05:29:06Z | cs.CL | Large language models (LLMs) often struggle with complex logical reasoning
due to logical inconsistencies and the inherent difficulty of such reasoning.
We use Lean, a theorem proving framework, to address these challenges. By
formalizing logical reasoning problems into theorems within Lean, we can solve
them by proving or disproving the corresponding theorems. This method reduces
the risk of logical inconsistencies with the help of Lean's symbolic solver. It
also enhances our ability to treat complex reasoning tasks by using Lean's
extensive library of theorem proofs. Our method achieves state-of-the-art
performance on the FOLIO dataset and achieves performance near this level on
ProofWriter. Notably, these results were accomplished by fine-tuning on fewer
than 100 in-domain samples for each dataset. | cs.CL | LeanReasoner: Boosting Complex Logical Reasoning with Lean |
2024-03-20T04:57:32Z | http://arxiv.org/pdf/2403.13301v1 | http://arxiv.org/abs/2403.13301v1 | Qihao Zhu, Leah Chong, Maria Yang, Jianxi Luo | 2024-03-20T04:57:32Z | cs.HC, cs.CL | In human-centered design, developing a comprehensive and in-depth
understanding of user experiences, i.e., empathic understanding, is paramount
for designing products that truly meet human needs. Nevertheless, accurately
comprehending the real underlying mental states of a large human population
remains a significant challenge today. This difficulty mainly arises from the
trade-off between depth and scale of user experience research: gaining in-depth
insights from a small group of users does not easily scale to a larger
population, and vice versa. This paper investigates the use of Large Language
Models (LLMs) for performing mental inference tasks, specifically inferring
users' underlying goals and fundamental psychological needs (FPNs). Baseline
and benchmark datasets were collected from human users and designers to develop
an empathic accuracy metric for measuring the mental inference performance of
LLMs. The empathic accuracy of inferring goals and FPNs of different LLMs with
varied zero-shot prompt engineering techniques are experimented against that of
human designers. Experimental results suggest that LLMs can infer and
understand the underlying goals and FPNs of users with performance comparable
to that of human designers, suggesting a promising avenue for enhancing the
scalability of empathic design approaches through the integration of advanced
artificial intelligence technologies. This work has the potential to
significantly augment the toolkit available to designers during human-centered
design, enabling the development of both large-scale and in-depth understanding
of users' experiences. | cs.HC | Reading Users' Minds from What They Say: An Investigation into LLM-based Empathic Mental Inference |
2024-03-20T03:14:54Z | http://arxiv.org/pdf/2403.13272v1 | http://arxiv.org/abs/2403.13272v1 | Md Towhidul Absar Chowdhury, Naveen Sharma, Ashiqur R. KhudaBukhsh | 2024-03-20T03:14:54Z | cs.CY, cs.CL, cs.SI | A community needs assessment is a tool used by non-profits and government
agencies to quantify the strengths and issues of a community, allowing them to
allocate their resources better. Such approaches are transitioning towards
leveraging social media conversations to analyze the needs of communities and
the assets already present within them. However, manual analysis of
exponentially increasing social media conversations is challenging. There is a
gap in the present literature in computationally analyzing how community
members discuss the strengths and needs of the community. To address this gap,
we introduce the task of identifying, extracting, and categorizing community
needs and assets from conversational data using sophisticated natural language
processing methods. To facilitate this task, we introduce the first dataset
about community needs and assets consisting of 3,511 conversations from Reddit,
annotated using crowdsourced workers. Using this dataset, we evaluate an
utterance-level classification model compared to sentiment classification and a
popular large language model (in a zero-shot setting), where we find that our
model outperforms both baselines at an F1 score of 94% compared to 49% and 61%
respectively. Furthermore, we observe through our study that conversations
about needs have negative sentiments and emotions, while conversations about
assets focus on location and entities. The dataset is available at
https://github.com/towhidabsar/CommunityNeeds. | cs.CY | Community Needs and Assets: A Computational Analysis of Community Conversations |
2024-03-20T03:07:50Z | http://arxiv.org/pdf/2403.13269v3 | http://arxiv.org/abs/2403.13269v3 | Zeyu Liu, Souvik Kundu, Anni Li, Junrui Wan, Lianghao Jiang, Peter Anthony Beerel | 2024-04-16T17:37:12Z | cs.CL, cs.AI, cs.LG | We present a novel Parameter-Efficient Fine-Tuning (PEFT) method, dubbed as
Adaptive Freezing of Low Rank Adaptation (AFLoRA). Specifically, for each
pre-trained frozen weight tensor, we add a parallel path of trainable low-rank
matrices, namely a down-projection and an up-projection matrix, each of which
is followed by a feature transformation vector. Based on a novel freezing
score, we the incrementally freeze these projection matrices during fine-tuning
to reduce the computation and alleviate over-fitting. Our experimental results
demonstrate that we can achieve state-of-the-art performance with an average
improvement of up to $0.85\%$ as evaluated on GLUE benchmark while yeilding up
to $9.5\times$ fewer average trainable parameters. While compared in terms of
runtime, AFLoRA can yield up to $1.86\times$ improvement as opposed to similar
PEFT alternatives. Besides the practical utility of our approach, we provide
insights on the trainability requirements of LoRA paths at different modules
and the freezing schedule for the different projection matrices. Code will be
released. | cs.CL | AFLoRA: Adaptive Freezing of Low Rank Adaptation in Parameter Efficient Fine-Tuning of Large Models |
2024-03-20T02:38:01Z | http://arxiv.org/pdf/2403.13257v2 | http://arxiv.org/abs/2403.13257v2 | Charles Goddard, Shamane Siriwardhana, Malikeh Ehghaghi, Luke Meyers, Vlad Karpukhin, Brian Benedict, Mark McQuade, Jacob Solawetz | 2024-03-21T03:13:30Z | cs.CL, cs.AI, cs.LG | The rapid expansion of the open-source language model landscape presents an
opportunity to merge the competencies of these model checkpoints by combining
their parameters. Advances in transfer learning, the process of fine-tuning
pretrained models for specific tasks, has resulted in the development of vast
amounts of task-specific models, typically specialized in individual tasks and
unable to utilize each other's strengths. Model merging facilitates the
creation of multitask models without the need for additional training, offering
a promising avenue for enhancing model performance and versatility. By
preserving the intrinsic capabilities of the original models, model merging
addresses complex challenges in AI - including the difficulties of catastrophic
forgetting and multitask learning. To support this expanding area of research,
we introduce MergeKit, a comprehensive, open-source library designed to
facilitate the application of model merging strategies. MergeKit offers an
extensible framework to efficiently merge models on any hardware, providing
utility to researchers and practitioners. To date, thousands of models have
been merged by the open-source community, leading to the creation of some of
the worlds most powerful open-source model checkpoints, as assessed by the Open
LLM Leaderboard. The library is accessible at
https://github.com/arcee-ai/MergeKit. | cs.CL | Arcee's MergeKit: A Toolkit for Merging Large Language Models |
2024-03-20T02:32:24Z | http://arxiv.org/pdf/2403.13253v1 | http://arxiv.org/abs/2403.13253v1 | Todd K Moon, Jacob H. Gunther | 2024-03-20T02:32:24Z | cs.CL, eess.AS | Over the years there has been ongoing interest in detecting authorship of a
text based on statistical properties of the text, such as by using occurrence
rates of noncontextual words. In previous work, these techniques have been
used, for example, to determine authorship of all of \emph{The Federalist
Papers}. Such methods may be useful in more modern times to detect fake or AI
authorship. Progress in statistical natural language parsers introduces the
possibility of using grammatical structure to detect authorship. In this paper
we explore a new possibility for detecting authorship using grammatical
structural information extracted using a statistical natural language parser.
This paper provides a proof of concept, testing author classification based on
grammatical structure on a set of "proof texts," The Federalist Papers and
Sanditon which have been as test cases in previous authorship detection
studies. Several features extracted from the statistical natural language
parser were explored: all subtrees of some depth from any level; rooted
subtrees of some depth, part of speech, and part of speech by level in the
parse tree. It was found to be helpful to project the features into a lower
dimensional space. Statistical experiments on these documents demonstrate that
information from a statistical parser can, in fact, assist in distinguishing
authors. | cs.CL | Document Author Classification Using Parsed Language Structure |
2024-03-20T02:29:09Z | http://arxiv.org/pdf/2403.13250v1 | http://arxiv.org/abs/2403.13250v1 | Huachuan Qiu, Shuai Zhang, Hongliang He, Anqi Li, Zhenzhong Lan | 2024-03-20T02:29:09Z | cs.CL | Pornographic content occurring in human-machine interaction dialogues can
cause severe side effects for users in open-domain dialogue systems. However,
research on detecting pornographic language within human-machine interaction
dialogues is an important subject that is rarely studied. To advance in this
direction, we introduce CensorChat, a dialogue monitoring dataset aimed at
detecting whether the dialogue session contains pornographic content. To this
end, we collect real-life human-machine interaction dialogues in the wild and
break them down into single utterances and single-turn dialogues, with the last
utterance spoken by the chatbot. We propose utilizing knowledge distillation of
large language models to annotate the dataset. Specifically, first, the raw
dataset is annotated by four open-source large language models, with the
majority vote determining the label. Second, we use ChatGPT to update the empty
label from the first step. Third, to ensure the quality of the validation and
test sets, we utilize GPT-4 for label calibration. If the current label does
not match the one generated by GPT-4, we employ a self-criticism strategy to
verify its correctness. Finally, to facilitate the detection of pornographic
text, we develop a series of text classifiers using a pseudo-labeled dataset.
Detailed data analysis demonstrates that leveraging knowledge distillation
techniques with large language models provides a practical and cost-efficient
method for developing pornographic text detectors. | cs.CL | Facilitating Pornographic Text Detection for Open-Domain Dialogue Systems via Knowledge Distillation of Large Language Models |
2024-03-20T02:15:55Z | http://arxiv.org/pdf/2403.13244v1 | http://arxiv.org/abs/2403.13244v1 | Peng Zhou, Jianmin Wang, Chunyan Li, Zixu Wang, Yiping Liu, Siqi Sun, Jianxin Lin, Longyue Wang, Xiangxiang Zeng | 2024-03-20T02:15:55Z | cs.CL, cs.AI | While various models and computational tools have been proposed for structure
and property analysis of molecules, generating molecules that conform to all
desired structures and properties remains a challenge. Here, we introduce a
multi-constraint molecular generation large language model, TSMMG, which, akin
to a student, incorporates knowledge from various small models and tools,
namely, the 'teachers'. To train TSMMG, we construct a large set of
text-molecule pairs by extracting molecular knowledge from these 'teachers',
enabling it to generate novel molecules that conform to the descriptions
through various text prompts. We experimentally show that TSMMG remarkably
performs in generating molecules meeting complex, natural language-described
property requirements across two-, three-, and four-constraint tasks, with an
average molecular validity of over 99% and success ratio of 88.08%, 65.27%, and
61.44%, respectively. The model also exhibits adaptability through zero-shot
testing, creating molecules that satisfy combinations of properties that have
not been encountered. It can comprehend text inputs with various language
styles, extending beyond the confines of outlined prompts, as confirmed through
empirical validation. Additionally, the knowledge distillation feature of TSMMG
contributes to the continuous enhancement of small models, while the innovative
approach to dataset construction effectively addresses the issues of data
scarcity and quality, which positions TSMMG as a promising tool in the domains
of drug discovery and materials science. Code is available at
https://github.com/HHW-zhou/TSMMG. | cs.CL | Instruction Multi-Constraint Molecular Generation Using a Teacher-Student Large Language Model |
2024-03-20T02:04:42Z | http://arxiv.org/pdf/2403.13240v1 | http://arxiv.org/abs/2403.13240v1 | Jacob Parnell, Inigo Jauregi Unanue, Massimo Piccardi | 2024-03-20T02:04:42Z | cs.CL | Cross-lingual summarization (XLS) generates summaries in a language different
from that of the input documents (e.g., English to Spanish), allowing speakers
of the target language to gain a concise view of their content. In the present
day, the predominant approach to this task is to take a performing, pretrained
multilingual language model (LM) and fine-tune it for XLS on the language pairs
of interest. However, the scarcity of fine-tuning samples makes this approach
challenging in some cases. For this reason, in this paper we propose revisiting
the summarize-and-translate pipeline, where the summarization and translation
tasks are performed in a sequence. This approach allows reusing the many,
publicly-available resources for monolingual summarization and translation,
obtaining a very competitive zero-shot performance. In addition, the proposed
pipeline is completely differentiable end-to-end, allowing it to take advantage
of few-shot fine-tuning, where available. Experiments over two contemporary and
widely adopted XLS datasets (CrossSum and WikiLingua) have shown the remarkable
zero-shot performance of the proposed approach, and also its strong few-shot
performance compared to an equivalent multilingual LM baseline, that the
proposed approach has been able to outperform in many languages with only 10%
of the fine-tuning samples. | cs.CL | SumTra: A Differentiable Pipeline for Few-Shot Cross-Lingual Summarization |
2024-03-20T01:46:06Z | http://arxiv.org/pdf/2403.13233v1 | http://arxiv.org/abs/2403.13233v1 | Shuaijiang Zhao, Xiaoquan Fang | 2024-03-20T01:46:06Z | cs.CL | In the era of flourishing large-scale models, the challenge of selecting and
optimizing datasets from the vast and complex sea of data, to enhance the
performance of large language models within the constraints of limited
computational resources, has become paramount. This paper details our solution
for the BetterMixture challenge, which focuses on the fine-tuning data mixing
for large language models. Our approach, which secured third place,
incorporates data deduplication, low-level and high-level quality filtering,
and diversity selection. The foundation of our solution is Ke-Data-Juicer, an
extension of Data-Juicer, demonstrating its robust capabilities in handling and
optimizing data for large language models. | cs.CL | Technical Report: Competition Solution For BetterMixture |
2024-03-20T00:22:38Z | http://arxiv.org/pdf/2403.13213v3 | http://arxiv.org/abs/2403.13213v3 | Khaoula Chehbouni, Megha Roshan, Emmanuel Ma, Futian Andrew Wei, Afaf Taik, Jackie CK Cheung, Golnoosh Farnadi | 2024-06-08T01:58:20Z | cs.LG, cs.CL, cs.CY | Recent progress in large language models (LLMs) has led to their widespread
adoption in various domains. However, these advancements have also introduced
additional safety risks and raised concerns regarding their detrimental impact
on already marginalized populations. Despite growing mitigation efforts to
develop safety safeguards, such as supervised safety-oriented fine-tuning and
leveraging safe reinforcement learning from human feedback, multiple concerns
regarding the safety and ingrained biases in these models remain. Furthermore,
previous work has demonstrated that models optimized for safety often display
exaggerated safety behaviors, such as a tendency to refrain from responding to
certain requests as a precautionary measure. As such, a clear trade-off between
the helpfulness and safety of these models has been documented in the
literature. In this paper, we further investigate the effectiveness of safety
measures by evaluating models on already mitigated biases. Using the case of
Llama 2 as an example, we illustrate how LLMs' safety responses can still
encode harmful assumptions. To do so, we create a set of non-toxic prompts,
which we then use to evaluate Llama models. Through our new taxonomy of LLMs
responses to users, we observe that the safety/helpfulness trade-offs are more
pronounced for certain demographic groups which can lead to quality-of-service
harms for marginalized populations. | cs.LG | From Representational Harms to Quality-of-Service Harms: A Case Study on Llama 2 Safety Safeguards |
2024-03-19T21:46:52Z | http://arxiv.org/pdf/2404.12394v1 | http://arxiv.org/abs/2404.12394v1 | Mohamed A. Allayla, Serkan Ayvaz | 2024-03-19T21:46:52Z | cs.LG, cs.CL, cs.SI | Online social media platforms have recently become integral to our society
and daily routines. Every day, users worldwide spend a couple of hours on such
platforms, expressing their sentiments and emotional state and contacting each
other. Analyzing such huge amounts of data from these platforms can provide a
clear insight into public sentiments and help detect their mental status. The
early identification of these health condition risks may assist in preventing
or reducing the number of suicide ideation and potentially saving people's
lives. The traditional techniques have become ineffective in processing such
streams and large-scale datasets. Therefore, the paper proposed a new
methodology based on a big data architecture to predict suicidal ideation from
social media content. The proposed approach provides a practical analysis of
social media data in two phases: batch processing and real-time streaming
prediction. The batch dataset was collected from the Reddit forum and used for
model building and training, while streaming big data was extracted using
Twitter streaming API and used for real-time prediction. After the raw data was
preprocessed, the extracted features were fed to multiple Apache Spark ML
classifiers: NB, LR, LinearSVC, DT, RF, and MLP. We conducted various
experiments using various feature-extraction techniques with different testing
scenarios. The experimental results of the batch processing phase showed that
the features extracted of (Unigram + Bigram) + CV-IDF with MLP classifier
provided high performance for classifying suicidal ideation, with an accuracy
of 93.47%, and then applied for real-time streaming prediction phase. | cs.LG | A Big Data Analytics System for Predicting Suicidal Ideation in Real-Time Based on Social Media Streaming Data |
2024-03-19T21:45:29Z | http://arxiv.org/pdf/2403.13169v2 | http://arxiv.org/abs/2403.13169v2 | Taiqi He, Kwanghee Choi, Lindia Tjuatja, Nathaniel R. Robinson, Jiatong Shi, Shinji Watanabe, Graham Neubig, David R. Mortensen, Lori Levin | 2024-06-06T03:37:32Z | cs.CL | Thousands of the world's languages are in danger of extinction--a tremendous
threat to cultural identities and human language diversity. Interlinear Glossed
Text (IGT) is a form of linguistic annotation that can support documentation
and resource creation for these languages' communities. IGT typically consists
of (1) transcriptions, (2) morphological segmentation, (3) glosses, and (4)
free translations to a majority language. We propose Wav2Gloss: a task in which
these four annotation components are extracted automatically from speech, and
introduce the first dataset to this end, Fieldwork: a corpus of speech with all
these annotations, derived from the work of field linguists, covering 37
languages, with standard formatting, and train/dev/test splits. We provide
various baselines to lay the groundwork for future research on IGT generation
from speech, such as end-to-end versus cascaded, monolingual versus
multilingual, and single-task versus multi-task approaches. | cs.CL | Wav2Gloss: Generating Interlinear Glossed Text from Speech |
2024-03-19T19:59:54Z | http://arxiv.org/pdf/2403.13130v1 | http://arxiv.org/abs/2403.13130v1 | Michele Resta, Davide Bacciu | 2024-03-19T19:59:54Z | cs.CL, cs.AI, cs.LG | Modern Neural Machine Translation systems exhibit strong performance in
several different languages and are constantly improving. Their ability to
learn continuously is, however, still severely limited by the catastrophic
forgetting issue. In this work, we leverage a key property of encoder-decoder
Transformers, i.e. their generative ability, to propose a novel approach to
continually learning Neural Machine Translation systems. We show how this can
effectively learn on a stream of experiences comprising different languages, by
leveraging a replay memory populated by using the model itself as a generator
of parallel sentences. We empirically demonstrate that our approach can
counteract catastrophic forgetting without requiring explicit memorization of
training data. Code will be publicly available upon publication. Code:
https://github.com/m-resta/sg-rep | cs.CL | Self-generated Replay Memories for Continual Neural Machine Translation |
2024-03-19T19:58:46Z | http://arxiv.org/pdf/2403.15465v1 | http://arxiv.org/abs/2403.15465v1 | Yuchao Li, Dimitri Bertsekas | 2024-03-19T19:58:46Z | cs.LG, cs.AI, cs.CL, cs.SY, eess.SY | In this paper we consider a transformer with an $n$-gram structure, such as
the one underlying ChatGPT. The transformer provides next word probabilities,
which can be used to generate word sequences. We consider methods for computing
word sequences that are highly likely, based on these probabilities. Computing
the optimal (i.e., most likely) word sequence starting with a given initial
state is an intractable problem, so we propose methods to compute highly likely
sequences of $N$ words in time that is a low order polynomial in $N$ and in the
vocabulary size of the $n$-gram. These methods are based on the rollout
approach from approximate dynamic programming, a form of single policy
iteration, which can improve the performance of any given heuristic policy. In
our case we use a greedy heuristic that generates as next word one that has the
highest probability. We show with analysis, examples, and computational
experimentation that our methods are capable of generating highly likely
sequences with a modest increase in computation over the greedy heuristic.
While our analysis and experiments are focused on Markov chains of the type
arising in transformer and ChatGPT-like models, our methods apply to general
finite-state Markov chains, and related inference applications of Hidden Markov
Models (HMM), where Viterbi decoding is used extensively. | cs.LG | Most Likely Sequence Generation for $n$-Grams, Transformers, HMMs, and Markov Chains, by Using Rollout Algorithms |
2024-03-19T19:27:23Z | http://arxiv.org/pdf/2403.13112v2 | http://arxiv.org/abs/2403.13112v2 | Bo-Ru Lu, Nikita Haduong, Chien-Yu Lin, Hao Cheng, Noah A. Smith, Mari Ostendorf | 2024-05-23T17:17:26Z | cs.CL | Transformer-based NLP models are powerful but have high computational costs
that limit deployment. Finetuned encoder-decoder models are popular in
specialized domains and can outperform larger more generalized decoder-only
models, such as GPT-4. We introduce a new configuration for encoder-decoder
models that improves efficiency on structured output and decomposable tasks
where multiple outputs are required for a single shared input. Our method,
prompt-in-decoder (PiD), encodes the input once and decodes the output in
parallel, boosting both training and inference efficiency by avoiding duplicate
input encoding and increasing the operational intensity (ratio of numbers of
arithmetic operation to memory access) of decoding process by sharing the input
key-value cache. We achieve computation reduction that roughly scales with the
number of subtasks, gaining up to 4.6x speed-up over state-of-the-art models
for dialogue state tracking, summarization, and question-answering tasks, with
comparable or better performance. | cs.CL | Efficient Encoder-Decoder Transformer Decoding for Decomposable Tasks |
2024-03-19T19:15:13Z | http://arxiv.org/pdf/2403.13107v2 | http://arxiv.org/abs/2403.13107v2 | M Manvith Prabhu, Haricharana Srinivasa, Anand Kumar M | 2024-07-02T03:35:53Z | cs.CL, cs.CY, cs.LG | This paper summarizes Team SCaLAR's work on SemEval-2024 Task 5: Legal
Argument Reasoning in Civil Procedure. To address this Binary Classification
task, which was daunting due to the complexity of the Legal Texts involved, we
propose a simple yet novel similarity and distance-based unsupervised approach
to generate labels. Further, we explore the Multi-level fusion of Legal-Bert
embeddings using ensemble features, including CNN, GRU, and LSTM. To address
the lengthy nature of Legal explanation in the dataset, we introduce T5-based
segment-wise summarization, which successfully retained crucial information,
enhancing the model's performance. Our unsupervised system witnessed a 20-point
increase in macro F1-score on the development set and a 10-point increase on
the test set, which is promising given its uncomplicated architecture. | cs.CL | Towards Unsupervised Question Answering System with Multi-level Summarization for Legal Text |
2024-03-19T19:13:22Z | http://arxiv.org/pdf/2403.13106v1 | http://arxiv.org/abs/2403.13106v1 | Divyansh Singhvi, Andrej Erkelens, Raghav Jain, Diganta Misra, Naomi Saphra | 2024-03-19T19:13:22Z | cs.LG, cs.AI, cs.CL, cs.CV | Measuring nonlinear feature interaction is an established approach to
understanding complex patterns of attribution in many models. In this paper, we
use Shapley Taylor interaction indices (STII) to analyze the impact of
underlying data structure on model representations in a variety of modalities,
tasks, and architectures. Considering linguistic structure in masked and
auto-regressive language models (MLMs and ALMs), we find that STII increases
within idiomatic expressions and that MLMs scale STII with syntactic distance,
relying more on syntax in their nonlinear structure than ALMs do. Our speech
model findings reflect the phonetic principal that the openness of the oral
cavity determines how much a phoneme varies based on its context. Finally, we
study image classifiers and illustrate that feature interactions intuitively
reflect object boundaries. Our wide range of results illustrates the benefits
of interdisciplinary work and domain expertise in interpretability research. | cs.LG | Knowing Your Nonlinearities: Shapley Interactions Reveal the Underlying Structure of Data |
2024-03-19T18:37:05Z | http://arxiv.org/pdf/2403.13089v1 | http://arxiv.org/abs/2403.13089v1 | Mengxian Lyu, Cheng Peng, Xiaohan Li, Patrick Balian, Jiang Bian, Yonghui Wu | 2024-03-19T18:37:05Z | cs.CL | Automatic text summarization (ATS) is an emerging technology to assist
clinicians in providing continuous and coordinated care. This study presents an
approach to summarize doctor-patient dialogues using generative large language
models (LLMs). We developed prompt-tuning algorithms to instruct generative
LLMs to summarize clinical text. We examined the prompt-tuning strategies, the
size of soft prompts, and the few-short learning ability of GatorTronGPT, a
generative clinical LLM developed using 277 billion clinical and general
English words with up to 20 billion parameters. We compared GatorTronGPT with a
previous solution based on fine-tuning of a widely used T5 model, using a
clinical benchmark dataset MTS-DIALOG. The experimental results show that the
GatorTronGPT- 20B model achieved the best performance on all evaluation
metrics. The proposed solution has a low computing cost as the LLM parameters
are not updated during prompt-tuning. This study demonstrates the efficiency of
generative clinical LLMs for clinical ATS through prompt tuning. | cs.CL | Automatic Summarization of Doctor-Patient Encounter Dialogues Using Large Language Model through Prompt Tuning |
2024-03-19T18:10:13Z | http://arxiv.org/pdf/2403.15464v1 | http://arxiv.org/abs/2403.15464v1 | Hejie Cui, Zhuocheng Shen, Jieyu Zhang, Hui Shao, Lianhui Qin, Joyce C. Ho, Carl Yang | 2024-03-19T18:10:13Z | cs.CL, cs.AI, cs.LG, cs.MA, J.3; I.2.7 | Electronic health records (EHRs) contain valuable patient data for
health-related prediction tasks, such as disease prediction. Traditional
approaches rely on supervised learning methods that require large labeled
datasets, which can be expensive and challenging to obtain. In this study, we
investigate the feasibility of applying Large Language Models (LLMs) to convert
structured patient visit data (e.g., diagnoses, labs, prescriptions) into
natural language narratives. We evaluate the zero-shot and few-shot performance
of LLMs using various EHR-prediction-oriented prompting strategies.
Furthermore, we propose a novel approach that utilizes LLM agents with
different roles: a predictor agent that makes predictions and generates
reasoning processes and a critic agent that analyzes incorrect predictions and
provides guidance for improving the reasoning of the predictor agent. Our
results demonstrate that with the proposed approach, LLMs can achieve decent
few-shot performance compared to traditional supervised learning methods in
EHR-based disease predictions, suggesting its potential for health-oriented
applications. | cs.CL | LLMs-based Few-Shot Disease Predictions using EHR: A Novel Approach Combining Predictive Agent Reasoning and Critical Agent Instruction |
2024-03-19T17:59:56Z | http://arxiv.org/pdf/2403.12968v1 | http://arxiv.org/abs/2403.12968v1 | Zhuoshi Pan, Qianhui Wu, Huiqiang Jiang, Menglin Xia, Xufang Luo, Jue Zhang, Qingwei Lin, Victor Rühle, Yuqing Yang, Chin-Yew Lin, H. Vicky Zhao, Lili Qiu, Dongmei Zhang | 2024-03-19T17:59:56Z | cs.CL, cs.LG | This paper focuses on task-agnostic prompt compression for better
generalizability and efficiency. Considering the redundancy in natural
language, existing approaches compress prompts by removing tokens or lexical
units according to their information entropy obtained from a causal language
model such as LLaMa-7B. The challenge is that information entropy may be a
suboptimal compression metric: (i) it only leverages unidirectional context and
may fail to capture all essential information needed for prompt compression;
(ii) it is not aligned with the prompt compression objective.
To address these issues, we propose a data distillation procedure to derive
knowledge from an LLM to compress prompts without losing crucial information,
and meantime, introduce an extractive text compression dataset. We formulate
prompt compression as a token classification problem to guarantee the
faithfulness of the compressed prompt to the original one, and use a
Transformer encoder as the base architecture to capture all essential
information for prompt compression from the full bidirectional context. Our
approach leads to lower latency by explicitly learning the compression
objective with smaller models such as XLM-RoBERTa-large and mBERT.
We evaluate our method on both in-domain and out-of-domain datasets,
including MeetingBank, LongBench, ZeroScrolls, GSM8K, and BBH. Despite its
small size, our model shows significant performance gains over strong baselines
and demonstrates robust generalization ability across different LLMs.
Additionally, our model is 3x-6x faster than existing prompt compression
methods, while accelerating the end-to-end latency by 1.6x-2.9x with
compression ratios of 2x-5x. | cs.CL | LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression |
2024-03-19T17:59:39Z | http://arxiv.org/pdf/2403.12964v1 | http://arxiv.org/abs/2403.12964v1 | Ce Zhang, Simon Stepputtis, Katia Sycara, Yaqi Xie | 2024-03-19T17:59:39Z | cs.CV, cs.CL | Recently, large-scale pre-trained Vision-Language Models (VLMs) have
demonstrated great potential in learning open-world visual representations, and
exhibit remarkable performance across a wide range of downstream tasks through
efficient fine-tuning. In this work, we innovatively introduce the concept of
dual learning into fine-tuning VLMs, i.e., we not only learn what an image is,
but also what an image isn't. Building on this concept, we introduce a novel
DualAdapter approach to enable dual-path adaptation of VLMs from both positive
and negative perspectives with only limited annotated samples. In the inference
stage, our DualAdapter performs unified predictions by simultaneously
conducting complementary positive selection and negative exclusion across
target classes, thereby enhancing the overall recognition accuracy of VLMs in
downstream tasks. Our extensive experimental results across 15 datasets
validate that the proposed DualAdapter outperforms existing state-of-the-art
methods on both few-shot learning and domain generalization tasks while
achieving competitive computational efficiency. Code is available at
https://github.com/zhangce01/DualAdapter. | cs.CV | Negative Yields Positive: Unified Dual-Path Adapter for Vision-Language Models |
2024-03-19T17:57:58Z | http://arxiv.org/pdf/2403.12958v1 | http://arxiv.org/abs/2403.12958v1 | Jeffrey Cheng, Marc Marone, Orion Weller, Dawn Lawrie, Daniel Khashabi, Benjamin Van Durme | 2024-03-19T17:57:58Z | cs.CL | Released Large Language Models (LLMs) are often paired with a claimed
knowledge cutoff date, or the dates at which training data was gathered. Such
information is crucial for applications where the LLM must provide up to date
information. However, this statement only scratches the surface: do all
resources in the training data share the same knowledge cutoff date? Does the
model's demonstrated knowledge for these subsets closely align to their cutoff
dates? In this work, we define the notion of an effective cutoff. This is
distinct from the LLM designer reported cutoff and applies separately to
sub-resources and topics. We propose a simple approach to estimate effective
cutoffs on the resource-level temporal alignment of an LLM by probing across
versions of the data. Using this analysis, we find that effective cutoffs often
differ from reported cutoffs. To understand the root cause of this observation,
we conduct a direct large-scale analysis on open pre-training datasets. Our
analysis reveals two reasons for these inconsistencies: (1) temporal biases of
CommonCrawl data due to non-trivial amounts of old data in new dumps and (2)
complications in LLM deduplication schemes involving semantic duplicates and
lexical near-duplicates. Overall, our results show that knowledge cutoffs are
not as simple as they have seemed and that care must be taken both by LLM
dataset curators as well as practitioners who seek to use information from
these models. | cs.CL | Dated Data: Tracing Knowledge Cutoffs in Large Language Models |
2024-03-19T17:43:08Z | http://arxiv.org/pdf/2403.12936v1 | http://arxiv.org/abs/2403.12936v1 | Joana Ribeiro de Faria, Huiyuan Xie, Felix Steffek | 2024-03-19T17:43:08Z | cs.CL, cs.AI | Court transcripts and judgments are rich repositories of legal knowledge,
detailing the intricacies of cases and the rationale behind judicial decisions.
The extraction of key information from these documents provides a concise
overview of a case, crucial for both legal experts and the public. With the
advent of large language models (LLMs), automatic information extraction has
become increasingly feasible and efficient. This paper presents a comprehensive
study on the application of GPT-4, a large language model, for automatic
information extraction from UK Employment Tribunal (UKET) cases. We
meticulously evaluated GPT-4's performance in extracting critical information
with a manual verification process to ensure the accuracy and relevance of the
extracted data. Our research is structured around two primary extraction tasks:
the first involves a general extraction of eight key aspects that hold
significance for both legal specialists and the general public, including the
facts of the case, the claims made, references to legal statutes, references to
precedents, general case outcomes and corresponding labels, detailed order and
remedies and reasons for the decision. The second task is more focused, aimed
at analysing three of those extracted features, namely facts, claims and
outcomes, in order to facilitate the development of a tool capable of
predicting the outcome of employment law disputes. Through our analysis, we
demonstrate that LLMs like GPT-4 can obtain high accuracy in legal information
extraction, highlighting the potential of LLMs in revolutionising the way legal
information is processed and utilised, offering significant implications for
legal research and practice. | cs.CL | Automatic Information Extraction From Employment Tribunal Judgements Using Large Language Models |
2024-03-19T17:28:51Z | http://arxiv.org/pdf/2403.12924v1 | http://arxiv.org/abs/2403.12924v1 | Grant Buster, Pavlo Pinchuk, Jacob Barrons, Ryan McKeever, Aaron Levine, Anthony Lopez | 2024-03-19T17:28:51Z | cs.CL | The recent growth in renewable energy development in the United States has
been accompanied by a simultaneous surge in renewable energy siting ordinances.
These zoning laws play a critical role in dictating the placement of wind and
solar resources that are critical for achieving low-carbon energy futures. In
this context, efficient access to and management of siting ordinance data
becomes imperative. The National Renewable Energy Laboratory (NREL) recently
introduced a public wind and solar siting database to fill this need. This
paper presents a method for harnessing Large Language Models (LLMs) to automate
the extraction of these siting ordinances from legal documents, enabling this
database to maintain accurate up-to-date information in the rapidly changing
energy policy landscape. A novel contribution of this research is the
integration of a decision tree framework with LLMs. Our results show that this
approach is 85 to 90% accurate with outputs that can be used directly in
downstream quantitative modeling. We discuss opportunities to use this work to
support similar large-scale policy research in the energy sector. By unlocking
new efficiencies in the extraction and analysis of legal documents using LLMs,
this study enables a path forward for automated large-scale energy policy
research. | cs.CL | Supporting Energy Policy Research with Large Language Models |
2024-03-19T17:21:29Z | http://arxiv.org/pdf/2403.12918v1 | http://arxiv.org/abs/2403.12918v1 | Sai Ashish Somayajula, Youwei Liang, Abhishek Singh, Li Zhang, Pengtao Xie | 2024-03-19T17:21:29Z | cs.CL, cs.AI, cs.LG | Pretrained Language Models (PLMs) have advanced Natural Language Processing
(NLP) tasks significantly, but finetuning PLMs on low-resource datasets poses
significant challenges such as instability and overfitting. Previous methods
tackle these issues by finetuning a strategically chosen subnetwork on a
downstream task, while keeping the remaining weights fixed to the pretrained
weights. However, they rely on a suboptimal criteria for sub-network selection,
leading to suboptimal solutions. To address these limitations, we propose a
regularization method based on attention-guided weight mixup for finetuning
PLMs. Our approach represents each network weight as a mixup of task-specific
weight and pretrained weight, controlled by a learnable attention parameter,
providing finer control over sub-network selection. Furthermore, we employ a
bi-level optimization (BLO) based framework on two separate splits of the
training dataset, improving generalization and combating overfitting. We
validate the efficacy of our proposed method through extensive experiments,
demonstrating its superiority over previous methods, particularly in the
context of finetuning PLMs on low-resource datasets. | cs.CL | Generalizable and Stable Finetuning of Pretrained Language Models on Low-Resource Texts |
2024-03-19T16:53:53Z | http://arxiv.org/pdf/2403.12900v1 | http://arxiv.org/abs/2403.12900v1 | Baolin Li, Yankai Jiang, Vijay Gadepally, Devesh Tiwari | 2024-03-19T16:53:53Z | cs.DC, cs.AI, cs.CL, cs.LG | The rapid advancement of Generative Artificial Intelligence (GenAI) across
diverse sectors raises significant environmental concerns, notably the carbon
emissions from their cloud and high performance computing (HPC) infrastructure.
This paper presents Sprout, an innovative framework designed to address these
concerns by reducing the carbon footprint of generative Large Language Model
(LLM) inference services. Sprout leverages the innovative concept of
"generation directives" to guide the autoregressive generation process, thereby
enhancing carbon efficiency. Our proposed method meticulously balances the need
for ecological sustainability with the demand for high-quality generation
outcomes. Employing a directive optimizer for the strategic assignment of
generation directives to user prompts and an original offline quality
evaluator, Sprout demonstrates a significant reduction in carbon emissions by
over 40% in real-world evaluations using the Llama2 LLM and global electricity
grid data. This research marks a critical step toward aligning AI technology
with sustainable practices, highlighting the potential for mitigating
environmental impacts in the rapidly expanding domain of generative artificial
intelligence. | cs.DC | Toward Sustainable GenAI using Generation Directives for Carbon-Friendly Large Language Model Inference |
2024-03-19T16:26:10Z | http://arxiv.org/pdf/2403.12881v1 | http://arxiv.org/abs/2403.12881v1 | Zehui Chen, Kuikun Liu, Qiuchen Wang, Wenwei Zhang, Jiangning Liu, Dahua Lin, Kai Chen, Feng Zhao | 2024-03-19T16:26:10Z | cs.CL | Open-sourced Large Language Models (LLMs) have achieved great success in
various NLP tasks, however, they are still far inferior to API-based models
when acting as agents. How to integrate agent ability into general LLMs becomes
a crucial and urgent problem. This paper first delivers three key observations:
(1) the current agent training corpus is entangled with both formats following
and agent reasoning, which significantly shifts from the distribution of its
pre-training data; (2) LLMs exhibit different learning speeds on the
capabilities required by agent tasks; and (3) current approaches have
side-effects when improving agent abilities by introducing hallucinations.
Based on the above findings, we propose Agent-FLAN to effectively Fine-tune
LANguage models for Agents. Through careful decomposition and redesign of the
training corpus, Agent-FLAN enables Llama2-7B to outperform prior best works by
3.5\% across various agent evaluation datasets. With comprehensively
constructed negative samples, Agent-FLAN greatly alleviates the hallucination
issues based on our established evaluation benchmark. Besides, it consistently
improves the agent capability of LLMs when scaling model sizes while slightly
enhancing the general capability of LLMs. The code will be available at
https://github.com/InternLM/Agent-FLAN. | cs.CL | Agent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models |
2024-03-19T16:06:10Z | http://arxiv.org/pdf/2403.12862v1 | http://arxiv.org/abs/2403.12862v1 | Minsu Kim, James Thorne | 2024-03-19T16:06:10Z | cs.CL | This paper investigates the inherent knowledge in language models from the
perspective of epistemological holism. The purpose of this paper is to explore
whether LLMs exhibit characteristics consistent with epistemological holism.
These characteristics suggest that core knowledge, such as general scientific
knowledge, each plays a specific role, serving as the foundation of our
knowledge system and being difficult to revise. To assess these traits related
to holism, we created a scientific reasoning dataset and examined the
epistemology of language models through three tasks: Abduction, Revision, and
Argument Generation. In the abduction task, the language models explained
situations while avoiding revising the core knowledge. However, in other tasks,
the language models were revealed not to distinguish between core and
peripheral knowledge, showing an incomplete alignment with holistic knowledge
principles. | cs.CL | Epistemology of Language Models: Do Language Models Have Holistic Knowledge? |
2024-03-19T15:07:22Z | http://arxiv.org/pdf/2403.12809v1 | http://arxiv.org/abs/2403.12809v1 | Zhixue Zhao, Nikolaos Aletras | 2024-03-19T15:07:22Z | cs.CL, cs.AI | In many real natural language processing application scenarios, practitioners
not only aim to maximize predictive performance but also seek faithful
explanations for the model predictions. Rationales and importance distribution
given by feature attribution methods (FAs) provide insights into how different
parts of the input contribute to a prediction. Previous studies have explored
how different factors affect faithfulness, mainly in the context of monolingual
English models. On the other hand, the differences in FA faithfulness between
multilingual and monolingual models have yet to be explored. Our extensive
experiments, covering five languages and five popular FAs, show that FA
faithfulness varies between multilingual and monolingual models. We find that
the larger the multilingual model, the less faithful the FAs are compared to
its counterpart monolingual models.Our further analysis shows that the
faithfulness disparity is potentially driven by the differences between model
tokenizers. Our code is available:
https://github.com/casszhao/multilingual-faith. | cs.CL | Comparing Explanation Faithfulness between Multilingual and Monolingual Fine-tuned Language Models |
2024-03-19T15:06:53Z | http://arxiv.org/pdf/2403.12805v1 | http://arxiv.org/abs/2403.12805v1 | Pierre Dognin, Jesus Rios, Ronny Luss, Inkit Padhi, Matthew D Riemer, Miao Liu, Prasanna Sattigeri, Manish Nagireddy, Kush R. Varshney, Djallel Bouneffouf | 2024-03-19T15:06:53Z | cs.AI, cs.CL | Developing value-aligned AI agents is a complex undertaking and an ongoing
challenge in the field of AI. Specifically within the domain of Large Language
Models (LLMs), the capability to consolidate multiple independently trained
dialogue agents, each aligned with a distinct moral value, into a unified
system that can adapt to and be aligned with multiple moral values is of
paramount importance. In this paper, we propose a system that does contextual
moral value alignment based on contextual aggregation. Here, aggregation is
defined as the process of integrating a subset of LLM responses that are best
suited to respond to a user input, taking into account features extracted from
the user's input. The proposed system shows better results in term of alignment
to human value compared to the state of the art. | cs.AI | Contextual Moral Value Alignment Through Context-Based Aggregation |
2024-03-19T15:01:14Z | http://arxiv.org/pdf/2403.12799v1 | http://arxiv.org/abs/2403.12799v1 | Mirza Alim Mutasodirin, Radityo Eko Prasojo | 2024-03-19T15:01:14Z | cs.CL, cs.AI | The parallelism of Transformer-based models comes at the cost of their input
max-length. Some studies proposed methods to overcome this limitation, but none
of them reported the effectiveness of summarization as an alternative. In this
study, we investigate the performance of document truncation and summarization
in text classification tasks. Each of the two was investigated with several
variations. This study also investigated how close their performances are to
the performance of full-text. We used a dataset of summarization tasks based on
Indonesian news articles (IndoSum) to do classification tests. This study shows
how the summaries outperform the majority of truncation method variations and
lose to only one. The best strategy obtained in this study is taking the head
of the document. The second is extractive summarization. This study explains
what happened to the result, leading to further research in order to exploit
the potential of document summarization as a shortening alternative. The code
and data used in this work are publicly available in
https://github.com/mirzaalimm/TruncationVsSummarization. | cs.CL | Investigating Text Shortening Strategy in BERT: Truncation vs Summarization |
2024-03-19T14:44:45Z | http://arxiv.org/pdf/2403.12776v1 | http://arxiv.org/abs/2403.12776v1 | Jiuhai Chen, Jonas Mueller | 2024-03-19T14:44:45Z | cs.CL | Large Language Models have become the de facto approach to
sequence-to-sequence text generation tasks, but for specialized tasks/domains,
a pretrained LLM lacks specific capabilities to produce accurate or
well-formatted responses. Supervised fine-tuning specializes a LLM by training
it on dataset of example prompts with target responses, but real-world data
tends to be noisy. While many fine-tuning algorithms exist, here we consider a
\emph{data-centric AI} perspective on LLM fine-tuning, studying how to
\emph{systematically} curate the training dataset to improve the LLM produced
via \emph{any} fine-tuning algorithm.
We introduce an automated data curation pipeline CLEAR (Confidence-based LLM
Evaluation And Rectification) for instruction tuning datasets, that can be used
with any LLM and fine-tuning procedure. CLEAR estimates which training data is
low-quality and either filters or corrects it. Automatically identifying which
data to filter or correct is done via LLM-derived confidence estimates, to
ensure only confident modifications to the dataset. Unlike existing data
curation techniques, CLEAR is a comprehensive framework that can improve a
dataset (and trained model outputs) without additional fine-tuning
computations. We don't assume access to a stronger LLM than the model being
fine-tuned (e.g.\ relying on GPT-4 when fine-tuning GPT-3.5), to see whether
CLEAR can meaningfully improve the capabilities of any LLM. Experiments reveal
that CLEAR consistently improves the performance of fine-tuned models across
many datasets and models (like GPT-3.5 and Llama2). | cs.CL | Automated Data Curation for Robust Language Model Fine-Tuning |
2024-03-19T14:12:54Z | http://arxiv.org/pdf/2403.12749v1 | http://arxiv.org/abs/2403.12749v1 | Siyao Peng, Zihang Sun, Huangyan Shan, Marie Kolm, Verena Blaschke, Ekaterina Artemova, Barbara Plank | 2024-03-19T14:12:54Z | cs.CL | Named Entity Recognition (NER) is a fundamental task to extract key
information from texts, but annotated resources are scarce for dialects. This
paper introduces the first dialectal NER dataset for German, BarNER, with 161K
tokens annotated on Bavarian Wikipedia articles (bar-wiki) and tweets
(bar-tweet), using a schema adapted from German CoNLL 2006 and GermEval. The
Bavarian dialect differs from standard German in lexical distribution,
syntactic construction, and entity information. We conduct in-domain,
cross-domain, sequential, and joint experiments on two Bavarian and three
German corpora and present the first comprehensive NER results on Bavarian.
Incorporating knowledge from the larger German NER (sub-)datasets notably
improves on bar-wiki and moderately on bar-tweet. Inversely, training first on
Bavarian contributes slightly to the seminal German CoNLL 2006 corpus.
Moreover, with gold dialect labels on Bavarian tweets, we assess multi-task
learning between five NER and two Bavarian-German dialect identification tasks
and achieve NER SOTA on bar-wiki. We substantiate the necessity of our
low-resource BarNER corpus and the importance of diversity in dialects, genres,
and topics in enhancing model performance. | cs.CL | Sebastian, Basti, Wastl?! Recognizing Named Entities in Bavarian Dialectal Data |
2024-03-19T14:11:20Z | http://arxiv.org/pdf/2403.13037v1 | http://arxiv.org/abs/2403.13037v1 | Rushi Qiang, Ruiyi Zhang, Pengtao Xie | 2024-03-19T14:11:20Z | cs.LG, cs.CL | Low-rank adaptation (LoRA) is a popular method for fine-tuning large-scale
pre-trained models in downstream tasks by learning low-rank incremental
matrices. Though LoRA and its variants effectively reduce the number of
trainable parameters compared to full fine-tuning methods, they often overfit
training data, resulting in sub-optimal generalization on test data. To address
this problem, we introduce BiLoRA, an overfitting-alleviating fine-tuning
approach based on bi-level optimization (BLO). BiLoRA employs pseudo singular
value decomposition to parameterize low-rank incremental matrices and splits
the training of pseudo singular vectors and values across two different subsets
of training data. This division, embedded within separate levels of the BLO
framework, mitigates the risk of overfitting to a single dataset. Tested on ten
datasets covering natural language understanding and generation tasks and
applied to various well-known large pre-trained models, BiLoRA significantly
outperforms LoRA methods and other fine-tuning approaches, with similar amounts
of trainable parameters. | cs.LG | BiLoRA: A Bi-level Optimization Framework for Overfitting-Resilient Low-Rank Adaptation of Large Pre-trained Models |
2024-03-19T14:07:28Z | http://arxiv.org/pdf/2403.12744v1 | http://arxiv.org/abs/2403.12744v1 | Zhenyu Wu, Chao Shen, Meng Jiang | 2024-03-19T14:07:28Z | cs.CL | Math word problem (MWP) solving requires generating a reasoning path based on
a given problem description that often contains irrelevant conditions. Existing
chain-of-thought (CoT) prompting methods elicited multi-step reasoning
abilities of large language models (LLMs) to solve MWPs. However, they were
seriously confused by the irrelevant conditions, resulting in low accuracy. In
this paper, we propose a novel approach named I$^3$C that instructs LLMs to
identify and ignore irrelevant conditions. It identifies a set of irrelevant
condition candidates that have a weak semantic relevance with the question.
Then it prompts LLMs to verify the irrelevant conditions. Lastly it instructs
the LLMs with the verification on relevant and irrelevant conditions to avoid
confusion and improve reasoning paths. Moreover, we propose to select (problem,
reasoning paths) pairs as demonstrations to enhance I$^3$C with few-shot
reasoning. We develop I$^3$C-Select that selects the most confusing problems
based on the semantic relevance measurement. We conduct extensive experiments
on eight MWP datasets. I$^3$C can be combined with any CoT prompting methods to
improve the performance of solving MWPs. Notably, with GPT-3.5-Turbo and
I$^3$C-Select, we achieve an accuracy of 96.0 and 94.1 on GSM-IC2-1K and
GSM-ICM-1K, respectively, significantly outperforming the state-of-the-art
few-shot prompting method Complex-CoT by +11.7 and +11.1. Our implementation is
made publicly available at https://wzy6642.github.io/I3C.github.io/. | cs.CL | Instructing Large Language Models to Identify and Ignore Irrelevant Conditions |
2024-03-19T13:30:47Z | http://arxiv.org/pdf/2403.12721v2 | http://arxiv.org/abs/2403.12721v2 | Nikola Ljubešić, Taja Kuzman | 2024-03-26T14:32:34Z | cs.CL | This paper presents a collection of highly comparable web corpora of
Slovenian, Croatian, Bosnian, Montenegrin, Serbian, Macedonian, and Bulgarian,
covering thereby the whole spectrum of official languages in the South Slavic
language space. The collection of these corpora comprises a total of 13 billion
tokens of texts from 26 million documents. The comparability of the corpora is
ensured by a comparable crawling setup and the usage of identical crawling and
post-processing technology. All the corpora were linguistically annotated with
the state-of-the-art CLASSLA-Stanza linguistic processing pipeline, and
enriched with document-level genre information via the Transformer-based
multilingual X-GENRE classifier, which further enhances comparability at the
level of linguistic annotation and metadata enrichment. The genre-focused
analysis of the resulting corpora shows a rather consistent distribution of
genres throughout the seven corpora, with variations in the most prominent
genre categories being well-explained by the economic strength of each language
community. A comparison of the distribution of genre categories across the
corpora indicates that web corpora from less developed countries primarily
consist of news articles. Conversely, web corpora from economically more
developed countries exhibit a smaller proportion of news content, with a
greater presence of promotional and opinionated texts. | cs.CL | CLASSLA-web: Comparable Web Corpora of South Slavic Languages Enriched with Linguistic and Genre Annotation |
2024-03-19T12:24:20Z | http://arxiv.org/pdf/2403.12678v1 | http://arxiv.org/abs/2403.12678v1 | Maksym Taranukhin, Sahithya Ravi, Gabor Lukacs, Evangelos Milios, Vered Shwartz | 2024-03-19T12:24:20Z | cs.CL, cs.AI | The Canadian air travel sector has seen a significant increase in flight
delays, cancellations, and other issues concerning passenger rights.
Recognizing this demand, we present a chatbot to assist passengers and educate
them about their rights. Our system breaks a complex user input into simple
queries which are used to retrieve information from a collection of documents
detailing air travel regulations. The most relevant passages from these
documents are presented along with links to the original documents and the
generated queries, enabling users to dissect and leverage the information for
their unique circumstances. The system successfully overcomes two predominant
challenges: understanding complex user inputs, and delivering accurate answers,
free of hallucinations, that passengers can rely on for making informed
decisions. A user study comparing the chatbot to a Google search demonstrated
the chatbot's usefulness and ease of use. Beyond the primary goal of providing
accurate and timely information to air passengers regarding their rights, we
hope that this system will also enable further research exploring the tradeoff
between the user-friendly conversational interface of chatbots and the accuracy
of retrieval systems. | cs.CL | Empowering Air Travelers: A Chatbot for Canadian Air Passenger Rights |
2024-03-19T12:21:20Z | http://arxiv.org/pdf/2403.12675v1 | http://arxiv.org/abs/2403.12675v1 | Dojun Park, Jiwoo Lee, Hyeyun Jeong, Seohyun Park, Sungeun Lee | 2024-03-19T12:21:20Z | cs.CL | The current evaluation of Large Language Models (LLMs) predominantly relies
on benchmarks focusing on their embedded knowledge by testing through
multiple-choice questions (MCQs), a format inherently suited for automated
evaluation. Our study extends this evaluation to explore LLMs' pragmatic
competence--a facet previously underexamined before the advent of sophisticated
LLMs, specifically in the context of Korean. We employ two distinct evaluation
setups: the conventional MCQ format, adapted for automatic evaluation, and
Open-Ended Questions (OEQs), assessed by human experts, to examine LLMs'
narrative response capabilities without predefined options. Our findings reveal
that GPT-4 excels, scoring 81.11 and 85.69 in the MCQ and OEQ setups,
respectively, with HyperCLOVA X, an LLM optimized for Korean, closely
following, especially in the OEQ setup, demonstrating a score of 81.56 with a
marginal difference of 4.13 points compared to GPT-4. Furthermore, while
few-shot learning strategies generally enhance LLM performance,
Chain-of-Thought (CoT) prompting introduces a bias toward literal
interpretations, hindering accurate pragmatic inference. Considering the
growing expectation for LLMs to understand and produce language that aligns
with human communicative norms, our findings emphasize the importance for
advancing LLMs' abilities to grasp and convey sophisticated meanings beyond
mere literal interpretations. | cs.CL | Pragmatic Competence Evaluation of Large Language Models for Korean |
2024-03-19T12:02:38Z | http://arxiv.org/pdf/2403.12666v1 | http://arxiv.org/abs/2403.12666v1 | Dojun Park, Sebastian Padó | 2024-03-19T12:02:38Z | cs.CL | Almost all frameworks for the manual or automatic evaluation of machine
translation characterize the quality of an MT output with a single number. An
exception is the Multidimensional Quality Metrics (MQM) framework which offers
a fine-grained ontology of quality dimensions for scoring (such as style,
fluency, accuracy, and terminology). Previous studies have demonstrated the
feasibility of MQM annotation but there are, to our knowledge, no computational
models that predict MQM scores for novel texts, due to a lack of resources. In
this paper, we address these shortcomings by (a) providing a 1200-sentence MQM
evaluation benchmark for the language pair English-Korean and (b) reframing MT
evaluation as the multi-task problem of simultaneously predicting several MQM
scores using SOTA language models, both in a reference-based MT evaluation
setup and a reference-free quality estimation (QE) setup. We find that
reference-free setup outperforms its counterpart in the style dimension while
reference-based models retain an edge regarding accuracy. Overall, RemBERT
emerges as the most promising model. Through our evaluation, we offer an
insight into the translation quality in a more fine-grained, interpretable
manner. | cs.CL | Multi-Dimensional Machine Translation Evaluation: Model Evaluation and Resource for Korean |
2024-03-19T11:49:03Z | http://arxiv.org/pdf/2403.15459v1 | http://arxiv.org/abs/2403.15459v1 | Bürki Audrey, Vasishth Shravan | 2024-03-19T11:49:03Z | cs.CL | With the pandemic, many experimental psychologists and linguists have started
to collect data over the internet (hereafter on-line data). The feasibility of
such experiments and the sample sizes required to achieve sufficient
statistical power in future experiments have to be assessed. This in turn
requires information on effect sizes and variability. In a series of analyses,
we compare response time data obtained in the same word production experiment
conducted in the lab and on-line. These analyses allow us to determine whether
the two settings differ in effect sizes, in the consistency of responses over
the course of the experiment, in the variability of average response times
across participants, in the magnitude of effect sizes across participants, or
in the amount of unexplained variability. We assess the impact of these
differences on the power of the design in a series of simulations. Our findings
temper the enthusiasm raised by previous studies and suggest that on-line
production studies might be feasible but at a non-negligible cost. The sample
sizes required to achieve sufficient power in on-line language production
studies come with a non-negligible increase in the amount of manual labour. | cs.CL | Assessing effect sizes, variability, and power in the on-line study of language production |
2024-03-19T11:36:53Z | http://arxiv.org/pdf/2403.15458v1 | http://arxiv.org/abs/2403.15458v1 | Daniel Fesalbon, Arvin De La Cruz, Marvin Mallari, Nelson Rodelas | 2024-03-19T11:36:53Z | cs.CL, cs.LG | Common problems in playing online mobile and computer games were related to
toxic behavior and abusive communication among players. Based on different
reports and studies, the study also discusses the impact of online hate speech
and toxicity on players' in-game performance and overall well-being. This study
investigates the capability of pre-trained language models to classify or
detect trash talk or toxic in-game messages The study employs and evaluates the
performance of pre-trained BERT and GPT language models in detecting toxicity
within in-game chats. Using publicly available APIs, in-game chat data from
DOTA 2 game matches were collected, processed, reviewed, and labeled as
non-toxic, mild (toxicity), and toxic. The study was able to collect around two
thousand in-game chats to train and test BERT (Base-uncased), BERT
(Large-uncased), and GPT-3 models. Based on the three models' state-of-the-art
performance, this study concludes pre-trained language models' promising
potential for addressing online hate speech and in-game insulting trash talk. | cs.CL | Fine-Tuning Pre-trained Language Models to Detect In-Game Trash Talks |
2024-03-19T10:11:14Z | http://arxiv.org/pdf/2403.12601v1 | http://arxiv.org/abs/2403.12601v1 | Chuang Liu, Renren Jin, Yuqi Ren, Deyi Xiong | 2024-03-19T10:11:14Z | cs.CL | Chinese Large Language Models (LLMs) have recently demonstrated impressive
capabilities across various NLP benchmarks and real-world applications.
However, the existing benchmarks for comprehensively evaluating these LLMs are
still insufficient, particularly in terms of measuring knowledge that LLMs
capture. Current datasets collect questions from Chinese examinations across
different subjects and educational levels to address this issue. Yet, these
benchmarks primarily focus on objective questions such as multiple-choice
questions, leading to a lack of diversity in question types. To tackle this
problem, we propose LHMKE, a Large-scale, Holistic, and Multi-subject Knowledge
Evaluation benchmark in this paper. LHMKE is designed to provide a
comprehensive evaluation of the knowledge acquisition capabilities of Chinese
LLMs. It encompasses 10,465 questions across 75 tasks covering 30 subjects,
ranging from primary school to professional certification exams. Notably, LHMKE
includes both objective and subjective questions, offering a more holistic
evaluation of the knowledge level of LLMs. We have assessed 11 Chinese LLMs
under the zero-shot setting, which aligns with real examinations, and compared
their performance across different subjects. We also conduct an in-depth
analysis to check whether GPT-4 can automatically score subjective predictions.
Our findings suggest that LHMKE is a challenging and advanced testbed for
Chinese LLMs. | cs.CL | LHMKE: A Large-scale Holistic Multi-subject Knowledge Evaluation Benchmark for Chinese Large Language Models |
2024-03-19T10:03:07Z | http://arxiv.org/pdf/2403.12596v1 | http://arxiv.org/abs/2403.12596v1 | Victor Carbune, Hassan Mansoor, Fangyu Liu, Rahul Aralikatte, Gilles Baechler, Jindong Chen, Abhanshu Sharma | 2024-03-19T10:03:07Z | cs.CL | Vision-language models (VLMs) are achieving increasingly strong performance
on multimodal tasks. However, reasoning capabilities remain limited
particularly for smaller VLMs, while those of large-language models (LLMs) have
seen numerous improvements. We propose a technique to transfer capabilities
from LLMs to VLMs. On the recently introduced ChartQA, our method obtains
state-of-the-art performance when applied on the PaLI3-5B VLM by
\citet{chen2023pali3}, while also enabling much better performance on PlotQA
and FigureQA.
We first improve the chart representation by continuing the pre-training
stage using an improved version of the chart-to-table translation task by
\citet{liu2023deplot}. We then propose constructing a 20x larger dataset than
the original training set. To improve general reasoning capabilities and
improve numerical operations, we synthesize reasoning traces using the table
representation of charts. Lastly, our model is fine-tuned using the multitask
loss introduced by \citet{hsieh2023distilling}.
Our variant ChartPaLI-5B outperforms even 10x larger models such as PaLIX-55B
without using an upstream OCR system, while keeping inference time constant
compared to the PaLI3-5B baseline. When rationales are further refined with a
simple program-of-thought prompt \cite{chen2023program}, our model outperforms
the recently introduced Gemini Ultra and GPT-4V. | cs.CL | Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs |
2024-03-19T09:45:33Z | http://arxiv.org/pdf/2403.12582v1 | http://arxiv.org/abs/2403.12582v1 | Xiang Li, Zhenyu Li, Chen Shi, Yong Xu, Qing Du, Mingkui Tan, Jun Huang, Wei Lin | 2024-03-19T09:45:33Z | cs.CL | The task of financial analysis primarily encompasses two key areas: stock
trend prediction and the corresponding financial question answering. Currently,
machine learning and deep learning algorithms (ML&DL) have been widely applied
for stock trend predictions, leading to significant progress. However, these
methods fail to provide reasons for predictions, lacking interpretability and
reasoning processes. Also, they can not integrate textual information such as
financial news or reports. Meanwhile, large language models (LLMs) have
remarkable textual understanding and generation ability. But due to the
scarcity of financial training datasets and limited integration with real-time
knowledge, LLMs still suffer from hallucinations and are unable to keep up with
the latest information. To tackle these challenges, we first release AlphaFin
datasets, combining traditional research datasets, real-time financial data,
and handwritten chain-of-thought (CoT) data. It has a positive impact on
training LLMs for completing financial analysis. We then use AlphaFin datasets
to benchmark a state-of-the-art method, called Stock-Chain, for effectively
tackling the financial analysis task, which integrates retrieval-augmented
generation (RAG) techniques. Extensive experiments are conducted to demonstrate
the effectiveness of our framework on financial analysis. | cs.CL | AlphaFin: Benchmarking Financial Analysis with Retrieval-Augmented Stock-Chain Framework |
2024-03-19T09:17:25Z | http://arxiv.org/pdf/2403.12563v1 | http://arxiv.org/abs/2403.12563v1 | Mirza Alim Mutasodirin, Radityo Eko Prasojo, Achmad F. Abka, Hanif Rasyidi | 2024-03-19T09:17:25Z | cs.CL, cs.AI | Many NLP researchers rely on free computational services, such as Google
Colab, to fine-tune their Transformer models, causing a limitation for
hyperparameter optimization (HPO) in long-text classification due to the method
having quadratic complexity and needing a bigger resource. In Indonesian, only
a few works were found on long-text classification using Transformers. Most
only use a small amount of data and do not report any HPO. In this study, using
18k news articles, we investigate which pretrained models are recommended to
use based on the output length of the tokenizer. We then compare some hacks to
shorten and enrich the sequences, which are the removals of stopwords,
punctuation, low-frequency words, and recurring words. To get a fair
comparison, we propose and run an efficient and dynamic HPO procedure that can
be done gradually on a limited resource and does not require a long-running
optimization library. Using the best hack found, we then compare 512, 256, and
128 tokens length. We find that removing stopwords while keeping punctuation
and low-frequency words is the best hack. Some of our setups manage to
outperform taking 512 first tokens using a smaller 128 or 256 first tokens
which manage to represent the same information while requiring less
computational resources. The findings could help developers to efficiently
pursue optimal performance of the models using limited resources. | cs.CL | Simple Hack for Transformers against Heavy Long-Text Classification on a Time- and Memory-Limited GPU Service |
2024-03-19T09:00:23Z | http://arxiv.org/pdf/2403.12556v1 | http://arxiv.org/abs/2403.12556v1 | Zhigang Chen, Benjia Zhou, Jun Li, Jun Wan, Zhen Lei, Ning Jiang, Quan Lu, Guoqing Zhao | 2024-03-19T09:00:23Z | cs.CL | Previous Sign Language Translation (SLT) methods achieve superior performance
by relying on gloss annotations. However, labeling high-quality glosses is a
labor-intensive task, which limits the further development of SLT. Although
some approaches work towards gloss-free SLT through jointly training the visual
encoder and translation network, these efforts still suffer from poor
performance and inefficient use of the powerful Large Language Model (LLM).
Most seriously, we find that directly introducing LLM into SLT will lead to
insufficient learning of visual representations as LLM dominates the learning
curve. To address these problems, we propose Factorized Learning assisted with
Large Language Model (FLa-LLM) for gloss-free SLT. Concretely, we factorize the
training process into two stages. In the visual initialing stage, we employ a
lightweight translation model after the visual encoder to pre-train the visual
encoder. In the LLM fine-tuning stage, we freeze the acquired knowledge in the
visual encoder and integrate it with a pre-trained LLM to inspire the LLM's
translation potential. This factorized training strategy proves to be highly
effective as evidenced by significant improvements achieved across three SLT
datasets which are all conducted under the gloss-free setting. | cs.CL | Factorized Learning Assisted with Large Language Model for Gloss-free Sign Language Translation |
2024-03-19T07:56:42Z | http://arxiv.org/pdf/2403.12526v1 | http://arxiv.org/abs/2403.12526v1 | Haochen Li, Di Geng | 2024-03-19T07:56:42Z | cs.CL | Events are essential components of speech and texts, describing the changes
in the state of entities. The event extraction task aims to identify and
classify events and find their participants according to event schemas.
Manually predefined event schemas have limited coverage and are hard to migrate
across domains. Therefore, the researchers propose Liberal Event Extraction
(LEE), which aims to extract events and discover event schemas simultaneously.
However, existing LEE models rely heavily on external language knowledge bases
and require the manual development of numerous rules for noise removal and
knowledge alignment, which is complex and laborious. To this end, we propose a
Prompt-based Graph Model for Liberal Event Extraction (PGLEE). Specifically, we
use a prompt-based model to obtain candidate triggers and arguments, and then
build heterogeneous event graphs to encode the structures within and between
events. Experimental results prove that our approach achieves excellent
performance with or without predefined event schemas, while the automatically
detected event schemas are proven high quality. | cs.CL | Prompt-based Graph Model for Joint Liberal Event Extraction and Event Schema Induction |
2024-03-19T07:50:32Z | http://arxiv.org/pdf/2403.12523v1 | http://arxiv.org/abs/2403.12523v1 | Haochen Li, Di Geng | 2024-03-19T07:50:32Z | cs.CL, cs.AI | Events describe the state changes of entities. In a document, multiple events
are connected by various relations (e.g., Coreference, Temporal, Causal, and
Subevent). Therefore, obtaining the connections between events through
Event-Event Relation Extraction (ERE) is critical to understand natural
language. There are two main problems in the current ERE works: a. Only
embeddings of the event triggers are used for event feature representation,
ignoring event arguments (e.g., time, place, person, etc.) and their structure
within the event. b. The interconnection between relations (e.g., temporal and
causal relations usually interact with each other ) is ignored. To solve the
above problems, this paper proposes a jointly multiple ERE framework called
GraphERE based on Graph-enhanced Event Embeddings. First, we enrich the event
embeddings with event argument and structure features by using static AMR
graphs and IE graphs; Then, to jointly extract multiple event relations, we use
Node Transformer and construct Task-specific Dynamic Event Graphs for each type
of relation. Finally, we used a multi-task learning strategy to train the whole
framework. Experimental results on the latest MAVEN-ERE dataset validate that
GraphERE significantly outperforms existing methods. Further analyses indicate
the effectiveness of the graph-enhanced event embeddings and the joint
extraction strategy. | cs.CL | GraphERE: Jointly Multiple Event-Event Relation Extraction via Graph-Enhanced Event Embeddings |
2024-03-19T07:25:02Z | http://arxiv.org/pdf/2403.13031v1 | http://arxiv.org/abs/2403.13031v1 | Zhuowen Yuan, Zidi Xiong, Yi Zeng, Ning Yu, Ruoxi Jia, Dawn Song, Bo Li | 2024-03-19T07:25:02Z | cs.CR, cs.AI, cs.CL, cs.LG | Recent advancements in Large Language Models (LLMs) have showcased remarkable
capabilities across various tasks in different domains. However, the emergence
of biases and the potential for generating harmful content in LLMs,
particularly under malicious inputs, pose significant challenges. Current
mitigation strategies, while effective, are not resilient under adversarial
attacks. This paper introduces Resilient Guardrails for Large Language Models
(RigorLLM), a novel framework designed to efficiently and effectively moderate
harmful and unsafe inputs and outputs for LLMs. By employing a multi-faceted
approach that includes energy-based training data augmentation through Langevin
dynamics, optimizing a safe suffix for inputs via minimax optimization, and
integrating a fusion-based model combining robust KNN with LLMs based on our
data augmentation, RigorLLM offers a robust solution to harmful content
moderation. Our experimental evaluations demonstrate that RigorLLM not only
outperforms existing baselines like OpenAI API and Perspective API in detecting
harmful content but also exhibits unparalleled resilience to jailbreaking
attacks. The innovative use of constrained optimization and a fusion-based
guardrail approach represents a significant step forward in developing more
secure and reliable LLMs, setting a new standard for content moderation
frameworks in the face of evolving digital threats. | cs.CR | RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content |
2024-03-19T07:07:13Z | http://arxiv.org/pdf/2403.12500v1 | http://arxiv.org/abs/2403.12500v1 | Shiki Sato, Reina Akama, Jun Suzuki, Kentaro Inui | 2024-03-19T07:07:13Z | cs.CL | Mitigating the generation of contradictory responses poses a substantial
challenge in dialogue response generation. The quality and quantity of
available contradictory response data play a vital role in suppressing these
contradictions, offering two significant benefits. First, having access to
large contradiction data enables a comprehensive examination of their
characteristics. Second, data-driven methods to mitigate contradictions may be
enhanced with large-scale contradiction data for training. Nevertheless, no
attempt has been made to build an extensive collection of model-generated
contradictory responses. In this paper, we build a large dataset of response
generation models' contradictions for the first time. Then, we acquire valuable
insights into the characteristics of model-generated contradictions through an
extensive analysis of the collected responses. Lastly, we also demonstrate how
this dataset substantially enhances the performance of data-driven
contradiction suppression methods. | cs.CL | A Large Collection of Model-generated Contradictory Responses for Consistency-aware Dialogue Systems |
2024-03-19T06:39:47Z | http://arxiv.org/pdf/2403.12482v2 | http://arxiv.org/abs/2403.12482v2 | Xudong Guo, Kaixuan Huang, Jiale Liu, Wenhui Fan, Natalia Vélez, Qingyun Wu, Huazheng Wang, Thomas L. Griffiths, Mengdi Wang | 2024-05-23T06:29:00Z | cs.AI, cs.CL, cs.CY, cs.MA | Large Language Models (LLMs) have emerged as integral tools for reasoning,
planning, and decision-making, drawing upon their extensive world knowledge and
proficiency in language-related tasks. LLMs thus hold tremendous potential for
natural language interaction within multi-agent systems to foster cooperation.
However, LLM agents tend to over-report and comply with any instruction, which
may result in information redundancy and confusion in multi-agent cooperation.
Inspired by human organizations, this paper introduces a framework that imposes
prompt-based organization structures on LLM agents to mitigate these problems.
Through a series of experiments with embodied LLM agents and human-agent
collaboration, our results highlight the impact of designated leadership on
team efficiency, shedding light on the leadership qualities displayed by LLM
agents and their spontaneous cooperative behaviors. Further, we harness the
potential of LLMs to propose enhanced organizational prompts, via a
Criticize-Reflect process, resulting in novel organization structures that
reduce communication costs and enhance team efficiency. | cs.AI | Embodied LLM Agents Learn to Cooperate in Organized Teams |
2024-03-19T06:39:23Z | http://arxiv.org/pdf/2403.15456v3 | http://arxiv.org/abs/2403.15456v3 | Seil Kang, Donghyun Kim, Junhyeok Kim, Hyo Kyung Lee, Seong Jae Hwang | 2024-03-29T04:38:51Z | cs.AI, cs.CL | Significant methodological strides have been made toward Chest X-ray (CXR)
understanding via modern vision-language models (VLMs), demonstrating
impressive Visual Question Answering (VQA) and CXR report generation abilities.
However, existing CXR understanding frameworks still possess several procedural
caveats. (1) Previous methods solely use CXR reports, which are insufficient
for comprehensive Visual Question Answering (VQA), especially when additional
health-related data like medication history and prior diagnoses are needed. (2)
Previous methods use raw CXR reports, which are often arbitrarily structured.
While modern language models can understand various text formats, restructuring
reports for clearer, organized anatomy-based information could enhance their
usefulness. (3) Current evaluation methods for CXR-VQA primarily emphasize
linguistic correctness, lacking the capability to offer nuanced assessments of
the generated answers. In this work, to address the aforementioned caveats, we
introduce WoLF, a Wide-scope Large Language Model Framework for CXR
understanding. To resolve (1), we capture multi-faceted records of patients,
which are utilized for accurate diagnoses in real-world clinical scenarios.
Specifically, we adopt the Electronic Health Records (EHR) to generate
instruction-following data suited for CXR understanding. Regarding (2), we
enhance report generation performance by decoupling knowledge in CXR reports
based on anatomical structure even within the attention step via masked
attention. To address (3), we introduce an AI-evaluation protocol optimized for
assessing the capabilities of LLM. Through extensive experimental validations,
WoLF demonstrates superior performance over other models on MIMIC-CXR in the
AI-evaluation arena about VQA (up to +9.47%p mean score) and by metrics about
report generation (+7.3%p BLEU-1). | cs.AI | WoLF: Wide-scope Large Language Model Framework for CXR Understanding |
2024-03-19T06:01:02Z | http://arxiv.org/pdf/2403.12469v1 | http://arxiv.org/abs/2403.12469v1 | Ojas Nimase, Sanghyun Hong | 2024-03-19T06:01:02Z | cs.CL, cs.LG | Sarcasm recognition is challenging because it needs an understanding of the
true intention, which is opposite to or different from the literal meaning of
the words. Prior work has addressed this challenge by developing a series of
methods that provide richer $contexts$, e.g., sentiment or cultural nuances, to
models. While shown to be effective individually, no study has systematically
evaluated their collective effectiveness. As a result, it remains unclear to
what extent additional contexts can improve sarcasm recognition. In this work,
we explore the improvements that existing methods bring by incorporating more
contexts into a model. To this end, we develop a framework where we can
integrate multiple contextual cues and test different approaches. In evaluation
with four approaches on three sarcasm recognition benchmarks, we achieve
existing state-of-the-art performances and also demonstrate the benefits of
sequentially adding more contexts. We also identify inherent drawbacks of using
more contexts, highlighting that in the pursuit of even better results, the
model may need to adopt societal biases. | cs.CL | When Do "More Contexts" Help with Sarcasm Recognition? |
2024-03-19T05:52:56Z | http://arxiv.org/pdf/2403.12468v1 | http://arxiv.org/abs/2403.12468v1 | Danqing Luo, Chen Zhang, Yan Zhang, Haizhou Li | 2024-03-19T05:52:56Z | cs.CL | Training or finetuning large-scale language models (LLMs) requires
substantial computation resources, motivating recent efforts to explore
parameter-efficient adaptation to downstream tasks. One approach is to treat
these models as black boxes and use forward passes (Inference APIs) to interact
with them. Current research focuses on adapting these black-box models to
downstream tasks using gradient-free prompt optimization, but this often
involves an expensive process of searching task-specific prompts. Therefore, we
are motivated to study black-box language model adaptation without prompt
search. Specifically, we introduce a label-enhanced cross-attention network
called CrossTune, which models the semantic relatedness between the input text
sequence and task-specific label descriptions. Its effectiveness is examined in
the context of few-shot text classification. To improve the generalization of
CrossTune, we utilize ChatGPT to generate additional training data through
in-context learning. A switch mechanism is implemented to exclude low-quality
ChatGPT-generated data. Through extensive experiments on seven benchmark text
classification datasets, we demonstrate that our proposed approach outperforms
the previous state-of-the-art gradient-free black-box tuning method by 5.7% on
average. Even without using ChatGPT-augmented data, CrossTune performs better
or comparably than previous black-box tuning methods, suggesting the
effectiveness of our approach. | cs.CL | CrossTune: Black-Box Few-Shot Classification with Label Enhancement |
2024-03-19T03:59:14Z | http://arxiv.org/pdf/2403.12416v3 | http://arxiv.org/abs/2403.12416v3 | Chong Ma, Hanqi Jiang, Wenting Chen, Yiwei Li, Zihao Wu, Xiaowei Yu, Zhengliang Liu, Lei Guo, Dajiang Zhu, Tuo Zhang, Dinggang Shen, Tianming Liu, Xiang Li | 2024-06-14T03:18:18Z | cs.CV, cs.CL, 68T07, I.2.0; I.4.0; I.5.4; I.7.0 | In the medical multi-modal frameworks, the alignment of cross-modality
features presents a significant challenge. However, existing works have learned
features that are implicitly aligned from the data, without considering the
explicit relationships in the medical context. This data-reliance may lead to
low generalization of the learned alignment relationships. In this work, we
propose the Eye-gaze Guided Multi-modal Alignment (EGMA) framework to harness
eye-gaze data for better alignment of medical visual and textual features. We
explore the natural auxiliary role of radiologists' eye-gaze data in aligning
medical images and text, and introduce a novel approach by using eye-gaze data,
collected synchronously by radiologists during diagnostic evaluations. We
conduct downstream tasks of image classification and image-text retrieval on
four medical datasets, where EGMA achieved state-of-the-art performance and
stronger generalization across different datasets. Additionally, we explore the
impact of varying amounts of eye-gaze data on model performance, highlighting
the feasibility and utility of integrating this auxiliary data into multi-modal
alignment framework. | cs.CV | Eye-gaze Guided Multi-modal Alignment for Medical Representation Learning |
2024-03-19T03:53:47Z | http://arxiv.org/pdf/2403.12413v1 | http://arxiv.org/abs/2403.12413v1 | Rahul Nadkarni, Yizhong Wang, Noah A. Smith | 2024-03-19T03:53:47Z | cs.CL | Language model-based instruction-following systems have lately shown
increasing performance on many benchmark tasks, demonstrating the capability of
adapting to a broad variety of instructions. However, such systems are often
not designed to be transparent about their limitations; a user may easily
prompt a model with an instruction without any idea of whether the responses
should be expected to be accurate, or if the system is even capable of
performing the task. We propose a third party performance prediction framework,
where a separate model is trained to predict the metric resulting from
evaluating an instruction-following system on a task while assuming access only
to its inputs and outputs at inference time. We perform this analysis with a
variety of both open and closed instruction-following models as well as
multiple performance predictors, and examine the effect of various factors such
as model size, number of training tasks, and prompt format. Our findings
indicate that third-party performance prediction is very challenging, and much
work remains in developing predictors that can automatically reveal the
limitations of modern instruction-following natural language processing
systems. | cs.CL | Third-Party Language Model Performance Prediction from Instruction |
2024-03-19T03:35:20Z | http://arxiv.org/pdf/2403.12408v1 | http://arxiv.org/abs/2403.12408v1 | Yifan Peng, Ilia Kulikov, Yilin Yang, Sravya Popuri, Hui Lu, Changhan Wang, Hongyu Gong | 2024-03-19T03:35:20Z | cs.CL, cs.SD, eess.AS | There have been emerging research interest and advances in speech-to-speech
translation (S2ST), translating utterances from one language to another. This
work proposes Multitask Speech Language Model (MSLM), which is a decoder-only
speech language model trained in a multitask setting. Without reliance on text
training data, our model is able to support multilingual S2ST with speaker
style preserved. | cs.CL | MSLM-S2ST: A Multitask Speech Language Model for Textless Speech-to-Speech Translation with Speaker Style Preservation |
2024-03-19T03:35:18Z | http://arxiv.org/pdf/2403.12407v1 | http://arxiv.org/abs/2403.12407v1 | Xiaoyu Qiu, Yuechen Wang, Jiaxin Shi, Wengang Zhou, Houqiang Li | 2024-03-19T03:35:18Z | cs.CL | Based on multilingual pre-trained models, cross-lingual transfer with prompt
learning has shown promising effectiveness, where soft prompt learned in a
source language is transferred to target languages for downstream tasks,
particularly in the low-resource scenario. To efficiently transfer soft prompt,
we propose a novel framework, Multilingual Prompt Translator (MPT), where a
multilingual prompt translator is introduced to properly process crucial
knowledge embedded in prompt by changing language knowledge while retaining
task knowledge. Concretely, we first train prompt in source language and employ
translator to translate it into target prompt. Besides, we extend an external
corpus as auxiliary data, on which an alignment task for predicted answer
probability is designed to convert language knowledge, thereby equipping target
prompt with multilingual knowledge. In few-shot settings on XNLI, MPT
demonstrates superiority over baselines by remarkable improvements. MPT is more
prominent compared with vanilla prompting when transferring to languages quite
distinct from source language. | cs.CL | Cross-Lingual Transfer for Natural Language Inference via Multilingual Prompt Translator |
2024-03-19T03:22:35Z | http://arxiv.org/pdf/2403.12403v2 | http://arxiv.org/abs/2403.12403v2 | Ayushi Nirmal, Amrita Bhattacharjee, Paras Sheth, Huan Liu | 2024-05-08T02:47:36Z | cs.CL, cs.AI, cs.LG | Although social media platforms are a prominent arena for users to engage in
interpersonal discussions and express opinions, the facade and anonymity
offered by social media may allow users to spew hate speech and offensive
content. Given the massive scale of such platforms, there arises a need to
automatically identify and flag instances of hate speech. Although several hate
speech detection methods exist, most of these black-box methods are not
interpretable or explainable by design. To address the lack of
interpretability, in this paper, we propose to use state-of-the-art Large
Language Models (LLMs) to extract features in the form of rationales from the
input text, to train a base hate speech classifier, thereby enabling faithful
interpretability by design. Our framework effectively combines the textual
understanding capabilities of LLMs and the discriminative power of
state-of-the-art hate speech classifiers to make these classifiers faithfully
interpretable. Our comprehensive evaluation on a variety of English language
social media hate speech datasets demonstrate: (1) the goodness of the
LLM-extracted rationales, and (2) the surprising retention of detector
performance even after training to ensure interpretability. All code and data
will be made available at https://github.com/AmritaBh/shield. | cs.CL | Towards Interpretable Hate Speech Detection using Large Language Model-extracted Rationales |
2024-03-19T03:22:28Z | http://arxiv.org/pdf/2403.12402v1 | http://arxiv.org/abs/2403.12402v1 | Yifan Peng, Ilia Kulikov, Yilin Yang, Sravya Popuri, Hui Lu, Changhan Wang, Hongyu Gong | 2024-03-19T03:22:28Z | cs.CL, cs.SD, eess.AS | Speech language models (LMs) are promising for high-quality speech synthesis
through in-context learning. A typical speech LM takes discrete semantic units
as content and a short utterance as prompt, and synthesizes speech which
preserves the content's semantics but mimics the prompt's style. However, there
is no systematic understanding on how the synthesized audio is controlled by
the prompt and content. In this work, we conduct an empirical study of the
widely used autoregressive (AR) and non-autoregressive (NAR) speech LMs and
provide insights into the prompt design and content semantic units. Our
analysis reveals that heterogeneous and nonstationary prompts hurt the audio
quality in contrast to the previous finding that longer prompts always lead to
better synthesis. Moreover, we find that the speaker style of the synthesized
audio is also affected by the content in addition to the prompt. We further
show that semantic units carry rich acoustic information such as pitch, tempo,
volume and speech emphasis, which might be leaked from the content to the
synthesized audio. | cs.CL | An Empirical Study of Speech Language Models for Prompt-Conditioned Speech Synthesis |
2024-03-19T03:00:03Z | http://arxiv.org/pdf/2403.12393v1 | http://arxiv.org/abs/2403.12393v1 | Yuan Gao, Yiheng Zhu, Yuanbin Cao, Yinzhi Zhou, Zhen Wu, Yujie Chen, Shenglan Wu, Haoyuan Hu, Xinyu Dai | 2024-03-19T03:00:03Z | cs.CL | Open Domain Multi-Hop Question Answering (ODMHQA) plays a crucial role in
Natural Language Processing (NLP) by aiming to answer complex questions through
multi-step reasoning over retrieved information from external knowledge
sources. Recently, Large Language Models (LLMs) have demonstrated remarkable
performance in solving ODMHQA owing to their capabilities including planning,
reasoning, and utilizing tools. However, LLMs may generate off-topic answers
when attempting to solve ODMHQA, namely the generated answers are irrelevant to
the original questions. This issue of off-topic answers accounts for
approximately one-third of incorrect answers, yet remains underexplored despite
its significance. To alleviate this issue, we propose the
Discriminate->Re-Compose->Re- Solve->Re-Decompose (Dr3) mechanism.
Specifically, the Discriminator leverages the intrinsic capabilities of LLMs to
judge whether the generated answers are off-topic. In cases where an off-topic
answer is detected, the Corrector performs step-wise revisions along the
reversed reasoning chain (Re-Compose->Re-Solve->Re-Decompose) until the final
answer becomes on-topic. Experimental results on the HotpotQA and
2WikiMultiHopQA datasets demonstrate that our Dr3 mechanism considerably
reduces the occurrence of off-topic answers in ODMHQA by nearly 13%, improving
the performance in Exact Match (EM) by nearly 3% compared to the baseline
method without the Dr3 mechanism. | cs.CL | Dr3: Ask Large Language Models Not to Give Off-Topic Answers in Open Domain Multi-Hop Question Answering |
2024-03-19T02:59:58Z | http://arxiv.org/pdf/2403.12392v1 | http://arxiv.org/abs/2403.12392v1 | Faisal Qarah | 2024-03-19T02:59:58Z | cs.CL, cs.AI | Arabic poetry, with its rich linguistic features and profound cultural
significance, presents a unique challenge to the Natural Language Processing
(NLP) field. The complexity of its structure and context necessitates advanced
computational models for accurate analysis. In this paper, we introduce
AraPoemBERT, an Arabic language model pretrained exclusively on Arabic poetry
text. To demonstrate the effectiveness of the proposed model, we compared
AraPoemBERT with 5 different Arabic language models on various NLP tasks
related to Arabic poetry. The new model outperformed all other models and
achieved state-of-the-art results in most of the downstream tasks. AraPoemBERT
achieved unprecedented accuracy in two out of three novel tasks: poet's gender
classification (99.34\% accuracy), and poetry sub-meter classification (97.79\%
accuracy). In addition, the model achieved an accuracy score in poems' rhyme
classification (97.73\% accuracy) which is almost equivalent to the best score
reported in this study. Moreover, the proposed model significantly outperformed
previous work and other comparative models in the tasks of poems' sentiment
analysis, achieving an accuracy of 78.95\%, and poetry meter classification
(99.03\% accuracy), while significantly expanding the scope of these two
problems. The dataset used in this study, contains more than 2.09 million
verses collected from online sources, each associated with various attributes
such as meter, sub-meter, poet, rhyme, and topic. The results demonstrate the
effectiveness of the proposed model in understanding and analyzing Arabic
poetry, achieving state-of-the-art results in several tasks and outperforming
previous works and other language models included in the study. AraPoemBERT
model is publicly available on \url{https://huggingface.co/faisalq}. | cs.CL | AraPoemBERT: A Pretrained Language Model for Arabic Poetry Analysis |
2024-03-19T02:52:58Z | http://arxiv.org/pdf/2403.12386v1 | http://arxiv.org/abs/2403.12386v1 | Pengchao Wu, Xuefeng Li, Jinghang Gu, Longhua Qian, Guodong Zhou | 2024-03-19T02:52:58Z | cs.CL, cs.AI | Biomedical event extraction is an information extraction task to obtain
events from biomedical text, whose targets include the type, the trigger, and
the respective arguments involved in an event. Traditional biomedical event
extraction usually adopts a pipelined approach, which contains trigger
identification, argument role recognition, and finally event construction
either using specific rules or by machine learning. In this paper, we propose
an n-ary relation extraction method based on the BERT pre-training model to
construct Binding events, in order to capture the semantic information about an
event's context and its participants. The experimental results show that our
method achieves promising results on the GE11 and GE13 corpora of the BioNLP
shared task with F1 scores of 63.14% and 59.40%, respectively. It demonstrates
that by significantly improving theperformance of Binding events, the overall
performance of the pipelined event extraction approach or even exceeds those of
current joint learning methods. | cs.CL | Pipelined Biomedical Event Extraction Rivaling Joint Learning |
2024-03-19T02:34:33Z | http://arxiv.org/pdf/2403.12374v1 | http://arxiv.org/abs/2403.12374v1 | Cheng Peng, Zehao Yu, Kaleb E Smith, Wei-Hsuan Lo-Ciganic, Jiang Bian, Yonghui Wu | 2024-03-19T02:34:33Z | cs.CL | The progress in natural language processing (NLP) using large language models
(LLMs) has greatly improved patient information extraction from clinical
narratives. However, most methods based on the fine-tuning strategy have
limited transfer learning ability for cross-domain applications. This study
proposed a novel approach that employs a soft prompt-based learning
architecture, which introduces trainable prompts to guide LLMs toward desired
outputs. We examined two types of LLM architectures, including encoder-only
GatorTron and decoder-only GatorTronGPT, and evaluated their performance for
the extraction of social determinants of health (SDoH) using a
cross-institution dataset from the 2022 n2c2 challenge and a cross-disease
dataset from the University of Florida (UF) Health. The results show that
decoder-only LLMs with prompt tuning achieved better performance in
cross-domain applications. GatorTronGPT achieved the best F1 scores for both
datasets, outperforming traditional fine-tuned GatorTron by 8.9% and 21.8% in a
cross-institution setting, and 5.5% and 14.5% in a cross-disease setting. | cs.CL | Improving Generalizability of Extracting Social Determinants of Health Using Large Language Models through Prompt-tuning |
2024-03-19T02:34:18Z | http://arxiv.org/pdf/2403.12373v3 | http://arxiv.org/abs/2403.12373v3 | Chi Hu, Yuan Ge, Xiangnan Ma, Hang Cao, Qiang Li, Yonghua Yang, Tong Xiao, Jingbo Zhu | 2024-03-22T06:18:54Z | cs.CL | Large Language Models (LLMs) have achieved impressive performance across
various reasoning tasks. However, even state-of-the-art LLMs such as ChatGPT
are prone to logical errors during their reasoning processes. Existing
solutions, such as deploying task-specific verifiers or voting over multiple
reasoning paths, either require extensive human annotations or fail in
scenarios with inconsistent responses. To address these challenges, we
introduce RankPrompt, a new prompting method that enables LLMs to self-rank
their responses without additional resources. RankPrompt breaks down the
ranking problem into a series of comparisons among diverse responses,
leveraging the inherent capabilities of LLMs to generate chains of comparison
as contextual exemplars. Our experiments across 11 arithmetic and commonsense
reasoning tasks show that RankPrompt significantly enhances the reasoning
performance of ChatGPT and GPT-4, with improvements of up to 13%. Moreover,
RankPrompt excels in LLM-based automatic evaluations for open-ended tasks,
aligning with human judgments 74% of the time in the AlpacaEval dataset. It
also exhibits robustness to variations in response order and consistency.
Collectively, our results validate RankPrompt as an effective method for
eliciting high-quality feedback from language models. | cs.CL | RankPrompt: Step-by-Step Comparisons Make Language Models Better Reasoners |
2024-03-19T02:25:29Z | http://arxiv.org/pdf/2403.12368v1 | http://arxiv.org/abs/2403.12368v1 | Xi Wang, Hongliang Dai, Shen Gao, Piji Li | 2024-03-19T02:25:29Z | cs.CL, cs.AI | The advancement of Large Language Models (LLMs) has led to significant
enhancements in the performance of chatbot systems. Many researchers have
dedicated their efforts to the development of bringing characteristics to
chatbots. While there have been commercial products for developing role-driven
chatbots using LLMs, it is worth noting that academic research in this area
remains relatively scarce. Our research focuses on investigating the
performance of LLMs in constructing Characteristic AI Agents by simulating
real-life individuals across different settings. Current investigations have
primarily focused on act on roles with simple profiles. In response to this
research gap, we create a benchmark for the characteristic AI agents task,
including dataset, techniques, and evaluation metrics. A dataset called
``Character100'' is built for this benchmark, comprising the most-visited
people on Wikipedia for language models to role-play. With the constructed
dataset, we conduct comprehensive assessment of LLMs across various settings.
In addition, we devise a set of automatic metrics for quantitative performance
evaluation. The experimental results underscore the potential directions for
further improvement in the capabilities of LLMs in constructing characteristic
AI agents. The benchmark is available at
https://github.com/nuaa-nlp/Character100. | cs.CL | Characteristic AI Agents via Large Language Models |
2024-03-18T23:48:33Z | http://arxiv.org/pdf/2403.12328v1 | http://arxiv.org/abs/2403.12328v1 | Cristiano Mesquita Garcia, Alessandro Lameiras Koerich, Alceu de Souza Britto Jr, Jean Paul Barddal | 2024-03-18T23:48:33Z | cs.LG, cs.CL, cs.IR | Systems and individuals produce data continuously. On the Internet, people
share their knowledge, sentiments, and opinions, provide reviews about services
and products, and so on. Automatically learning from these textual data can
provide insights to organizations and institutions, thus preventing financial
impacts, for example. To learn from textual data over time, the machine
learning system must account for concept drift. Concept drift is a frequent
phenomenon in real-world datasets and corresponds to changes in data
distribution over time. For instance, a concept drift occurs when sentiments
change or a word's meaning is adjusted over time. Although concept drift is
frequent in real-world applications, benchmark datasets with labeled drifts are
rare in the literature. To bridge this gap, this paper provides four textual
drift generation methods to ease the production of datasets with labeled
drifts. These methods were applied to Yelp and Airbnb datasets and tested using
incremental classifiers respecting the stream mining paradigm to evaluate their
ability to recover from the drifts. Results show that all methods have their
performance degraded right after the drifts, and the incremental SVM is the
fastest to run and recover the previous performance levels regarding accuracy
and Macro F1-Score. | cs.LG | Methods for Generating Drift in Text Streams |
2024-03-18T23:41:52Z | http://arxiv.org/pdf/2403.15455v1 | http://arxiv.org/abs/2403.15455v1 | Cristiano Mesquita Garcia, Alessandro Lameiras Koerich, Alceu de Souza Britto Jr, Jean Paul Barddal | 2024-03-18T23:41:52Z | cs.CL, cs.LG | The proliferation of textual data on the Internet presents a unique
opportunity for institutions and companies to monitor public opinion about
their services and products. Given the rapid generation of such data, the text
stream mining setting, which handles sequentially arriving, potentially
infinite text streams, is often more suitable than traditional batch learning.
While pre-trained language models are commonly employed for their high-quality
text vectorization capabilities in streaming contexts, they face challenges
adapting to concept drift - the phenomenon where the data distribution changes
over time, adversely affecting model performance. Addressing the issue of
concept drift, this study explores the efficacy of seven text sampling methods
designed to selectively fine-tune language models, thereby mitigating
performance degradation. We precisely assess the impact of these methods on
fine-tuning the SBERT model using four different loss functions. Our
evaluation, focused on Macro F1-score and elapsed time, employs two text stream
datasets and an incremental SVM classifier to benchmark performance. Our
findings indicate that Softmax loss and Batch All Triplets loss are
particularly effective for text stream classification, demonstrating that
larger sample sizes generally correlate with improved macro F1-scores. Notably,
our proposed WordPieceToken ratio sampling method significantly enhances
performance with the identified loss functions, surpassing baseline results. | cs.CL | Improving Sampling Methods for Fine-tuning SentenceBERT in Text Streams |
2024-03-18T23:22:50Z | http://arxiv.org/pdf/2403.15454v2 | http://arxiv.org/abs/2403.15454v2 | Mahdi Rezapour | 2024-03-28T21:26:39Z | cs.CL, stat.AP | In this study, we explore the application of transformer-based models for
emotion classification on text data. We train and evaluate several pre-trained
transformer models, on the Emotion dataset using different variants of
transformers. The paper also analyzes some factors that in-fluence the
performance of the model, such as the fine-tuning of the transformer layer, the
trainability of the layer, and the preprocessing of the text data. Our analysis
reveals that commonly applied techniques like removing punctuation and stop
words can hinder model performance. This might be because transformers strength
lies in understanding contextual relationships within text. Elements like
punctuation and stop words can still convey sentiment or emphasis and removing
them might disrupt this context. | cs.CL | Emotion Detection with Transformers: A Comparative Study |
2024-03-18T23:21:37Z | http://arxiv.org/pdf/2403.12316v1 | http://arxiv.org/abs/2403.12316v1 | Chuang Liu, Linhao Yu, Jiaxuan Li, Renren Jin, Yufei Huang, Ling Shi, Junhui Zhang, Xinmeng Ji, Tingting Cui, Tao Liu, Jinwang Song, Hongying Zan, Sun Li, Deyi Xiong | 2024-03-18T23:21:37Z | cs.CL | The rapid development of Chinese large language models (LLMs) poses big
challenges for efficient LLM evaluation. While current initiatives have
introduced new benchmarks or evaluation platforms for assessing Chinese LLMs,
many of these focus primarily on capabilities, usually overlooking potential
alignment and safety issues. To address this gap, we introduce OpenEval, an
evaluation testbed that benchmarks Chinese LLMs across capability, alignment
and safety. For capability assessment, we include 12 benchmark datasets to
evaluate Chinese LLMs from 4 sub-dimensions: NLP tasks, disciplinary knowledge,
commonsense reasoning and mathematical reasoning. For alignment assessment,
OpenEval contains 7 datasets that examines the bias, offensiveness and
illegalness in the outputs yielded by Chinese LLMs. To evaluate safety,
especially anticipated risks (e.g., power-seeking, self-awareness) of advanced
LLMs, we include 6 datasets. In addition to these benchmarks, we have
implemented a phased public evaluation and benchmark update strategy to ensure
that OpenEval is in line with the development of Chinese LLMs or even able to
provide cutting-edge benchmark datasets to guide the development of Chinese
LLMs. In our first public evaluation, we have tested a range of Chinese LLMs,
spanning from 7B to 72B parameters, including both open-source and proprietary
models. Evaluation results indicate that while Chinese LLMs have shown
impressive performance in certain tasks, more attention should be directed
towards broader aspects such as commonsense reasoning, alignment, and safety. | cs.CL | OpenEval: Benchmarking Chinese LLMs across Capability, Alignment and Safety |
2024-03-18T22:39:03Z | http://arxiv.org/pdf/2403.12297v1 | http://arxiv.org/abs/2403.12297v1 | Maria Mahbub, Gregory M. Dams, Sudarshan Srinivasan, Caitlin Rizy, Ioana Danciu, Jodie Trafton, Kathryn Knight | 2024-03-18T22:39:03Z | cs.CL, cs.AI | Substance use disorder (SUD) poses a major concern due to its detrimental
effects on health and society. SUD identification and treatment depend on a
variety of factors such as severity, co-determinants (e.g., withdrawal
symptoms), and social determinants of health. Existing diagnostic coding
systems used by American insurance providers, like the International
Classification of Diseases (ICD-10), lack granularity for certain diagnoses,
but clinicians will add this granularity (as that found within the Diagnostic
and Statistical Manual of Mental Disorders classification or DSM-5) as
supplemental unstructured text in clinical notes. Traditional natural language
processing (NLP) methods face limitations in accurately parsing such diverse
clinical language. Large Language Models (LLMs) offer promise in overcoming
these challenges by adapting to diverse language patterns. This study
investigates the application of LLMs for extracting severity-related
information for various SUD diagnoses from clinical notes. We propose a
workflow employing zero-shot learning of LLMs with carefully crafted prompts
and post-processing techniques. Through experimentation with Flan-T5, an
open-source LLM, we demonstrate its superior recall compared to the rule-based
approach. Focusing on 11 categories of SUD diagnoses, we show the effectiveness
of LLMs in extracting severity information, contributing to improved risk
assessment and treatment planning for SUD patients. | cs.CL | Leveraging Large Language Models to Extract Information on Substance Use Disorder Severity from Clinical Notes: A Zero-shot Learning Approach |
2024-03-18T22:33:51Z | http://arxiv.org/pdf/2403.12294v1 | http://arxiv.org/abs/2403.12294v1 | Elliot Murphy, Jill de Villiers, Sofia Lucero Morales | 2024-03-18T22:33:51Z | cs.CL | In this study we compared how well DALL-E 2 visually represented the meaning
of linguistic prompts also given to young children in comprehension tests.
Sentences representing fundamental components of grammatical knowledge were
selected from assessment tests used with several hundred English-speaking
children aged 2-7 years for whom we had collected original item-level data.
DALL-E 2 was given these prompts five times to generate 20 cartoons per item,
for 9 adult judges to score. Results revealed no conditions in which DALL-E
2-generated images that matched the semantic accuracy of children, even at the
youngest age (2 years). DALL-E 2 failed to assign the appropriate roles in
reversible forms; it failed on negation despite an easier contrastive prompt
than the children received; it often assigned the adjective to the wrong noun;
it ignored implicit agents in passives. This work points to a clear absence of
compositional sentence representations for DALL-E 2. | cs.CL | A Comparative Investigation of Compositional Syntax and Semantics in DALL-E 2 |
2024-03-18T22:11:00Z | http://arxiv.org/pdf/2403.12285v1 | http://arxiv.org/abs/2403.12285v1 | Thanos Konstantinidis, Giorgos Iacovides, Mingxue Xu, Tony G. Constantinides, Danilo Mandic | 2024-03-18T22:11:00Z | cs.CL, cs.LG, q-fin.ST, q-fin.TR | There are multiple sources of financial news online which influence market
movements and trader's decisions. This highlights the need for accurate
sentiment analysis, in addition to having appropriate algorithmic trading
techniques, to arrive at better informed trading decisions. Standard lexicon
based sentiment approaches have demonstrated their power in aiding financial
decisions. However, they are known to suffer from issues related to context
sensitivity and word ordering. Large Language Models (LLMs) can also be used in
this context, but they are not finance-specific and tend to require significant
computational resources. To facilitate a finance specific LLM framework, we
introduce a novel approach based on the Llama 2 7B foundational model, in order
to benefit from its generative nature and comprehensive language manipulation.
This is achieved by fine-tuning the Llama2 7B model on a small portion of
supervised financial sentiment analysis data, so as to jointly handle the
complexities of financial lexicon and context, and further equipping it with a
neural network based decision mechanism. Such a generator-classifier scheme,
referred to as FinLlama, is trained not only to classify the sentiment valence
but also quantify its strength, thus offering traders a nuanced insight into
financial news articles. Complementing this, the implementation of
parameter-efficient fine-tuning through LoRA optimises trainable parameters,
thus minimising computational and memory requirements, without sacrificing
accuracy. Simulation results demonstrate the ability of the proposed FinLlama
to provide a framework for enhanced portfolio management decisions and
increased market returns. These results underpin the ability of FinLlama to
construct high-return portfolios which exhibit enhanced resilience, even during
volatile periods and unpredictable market events. | cs.CL | FinLlama: Financial Sentiment Classification for Algorithmic Trading Applications |
2024-03-18T20:50:26Z | http://arxiv.org/pdf/2403.12244v1 | http://arxiv.org/abs/2403.12244v1 | Patanjali Bhamidipati, Advaith Malladi, Manish Shrivastava, Radhika Mamidi | 2024-03-18T20:50:26Z | cs.CL | In recent studies, the extensive utilization of large language models has
underscored the importance of robust evaluation methodologies for assessing
text generation quality and relevance to specific tasks. This has revealed a
prevalent issue known as hallucination, an emergent condition in the model
where generated text lacks faithfulness to the source and deviates from the
evaluation criteria. In this study, we formally define hallucination and
propose a framework for its quantitative detection in a zero-shot setting,
leveraging our definition and the assumption that model outputs entail task and
sample specific inputs. In detecting hallucinations, our solution achieves an
accuracy of 0.78 in a model-aware setting and 0.61 in a model-agnostic setting.
Notably, our solution maintains computational efficiency, requiring far less
computational resources than other SOTA approaches, aligning with the trend
towards lightweight and compressed models. | cs.CL | Zero-Shot Multi-task Hallucination Detection |
2024-03-18T20:47:10Z | http://arxiv.org/pdf/2403.12242v2 | http://arxiv.org/abs/2403.12242v2 | Bang Nguyen, Mengxia Yu, Yun Huang, Meng Jiang | 2024-06-17T15:33:37Z | cs.CL, cs.AI, cs.LG | Reference-based metrics such as BLEU and BERTScore are widely used to
evaluate question generation (QG). In this study, on QG benchmarks such as
SQuAD and HotpotQA, we find that using human-written references cannot
guarantee the effectiveness of the reference-based metrics. Most QG benchmarks
have only one reference; we replicated the annotation process and collect
another reference. A good metric was expected to grade a human-validated
question no worse than generated questions. However, the results of
reference-based metrics on our newly collected reference disproved the metrics
themselves. We propose a reference-free metric consisted of multi-dimensional
criteria such as naturalness, answerability, and complexity, utilizing large
language models. These criteria are not constrained to the syntactic or
semantic of a single reference question, and the metric does not require a
diverse set of references. Experiments reveal that our metric accurately
distinguishes between high-quality questions and flawed ones, and achieves
state-of-the-art alignment with human judgment. | cs.CL | Reference-based Metrics Disprove Themselves in Question Generation |
2024-03-18T20:10:44Z | http://arxiv.org/pdf/2403.15453v1 | http://arxiv.org/abs/2403.15453v1 | Yifan Ding, Michael Yankoski, Tim Weninger | 2024-03-18T20:10:44Z | cs.CL, cs.AI, cs.IR | Information Extraction refers to a collection of tasks within Natural
Language Processing (NLP) that identifies sub-sequences within text and their
labels. These tasks have been used for many years to link extract relevant
information and to link free text to structured data. However, the
heterogeneity among information extraction tasks impedes progress in this area.
We therefore offer a unifying perspective centered on what we define to be
spans in text. We then re-orient these seemingly incongruous tasks into this
unified perspective and then re-present the wide assortment of information
extraction tasks as variants of the same basic Span-Oriented Information
Extraction task. | cs.CL | Span-Oriented Information Extraction -- A Unifying Perspective on Information Extraction |
2024-03-18T19:53:56Z | http://arxiv.org/pdf/2403.12212v1 | http://arxiv.org/abs/2403.12212v1 | Ramon Abilio, Guilherme Palermo Coelho, Ana Estela Antunes da Silva | 2024-03-18T19:53:56Z | cs.CL, cs.AI, cs.LG, 68T50 | Named Entity Recognition (NER) is a Natural Language Processing technique for
extracting information from textual documents. However, much of the existing
research on NER has been centered around English-language documents, leaving a
gap in the availability of datasets tailored to the financial domain in
Portuguese. This study addresses the need for NER within the financial domain,
focusing on Portuguese-language texts extracted from earnings call
transcriptions of Brazilian banks. By curating a comprehensive dataset
comprising 384 transcriptions and leveraging weak supervision techniques for
annotation, we evaluate the performance of monolingual models trained on
Portuguese (BERTimbau and PTT5) and multilingual models (mBERT and mT5).
Notably, we introduce a novel approach that reframes the token classification
task as a text generation problem, enabling fine-tuning and evaluation of T5
models. Following the fine-tuning of the models, we conduct an evaluation on
the test dataset, employing performance and error metrics. Our findings reveal
that BERT-based models consistently outperform T5-based models. Furthermore,
while the multilingual models exhibit comparable macro F1-scores, BERTimbau
demonstrates superior performance over PTT5. A manual analysis of sentences
generated by PTT5 and mT5 unveils a degree of similarity ranging from 0.89 to
1.0, between the original and generated sentences. However, critical errors
emerge as both models exhibit discrepancies, such as alterations to monetary
and percentage values, underscoring the importance of accuracy and consistency
in the financial domain. Despite these challenges, PTT5 and mT5 achieve
impressive macro F1-scores of 98.52% and 98.85%, respectively, with our
proposed approach. Furthermore, our study sheds light on notable disparities in
memory and time consumption for inference across the models. | cs.CL | Evaluating Named Entity Recognition: Comparative Analysis of Mono- and Multilingual Transformer Models on Brazilian Corporate Earnings Call Transcriptions |
2024-03-18T18:45:28Z | http://arxiv.org/pdf/2403.12173v1 | http://arxiv.org/abs/2403.12173v1 | Mengting Wan, Tara Safavi, Sujay Kumar Jauhar, Yujin Kim, Scott Counts, Jennifer Neville, Siddharth Suri, Chirag Shah, Ryen W White, Longqi Yang, Reid Andersen, Georg Buscher, Dhruv Joshi, Nagu Rangan | 2024-03-18T18:45:28Z | cs.CL, cs.AI, cs.IR | Transforming unstructured text into structured and meaningful forms,
organized by useful category labels, is a fundamental step in text mining for
downstream analysis and application. However, most existing methods for
producing label taxonomies and building text-based label classifiers still rely
heavily on domain expertise and manual curation, making the process expensive
and time-consuming. This is particularly challenging when the label space is
under-specified and large-scale data annotations are unavailable. In this
paper, we address these challenges with Large Language Models (LLMs), whose
prompt-based interface facilitates the induction and use of large-scale pseudo
labels. We propose TnT-LLM, a two-phase framework that employs LLMs to automate
the process of end-to-end label generation and assignment with minimal human
effort for any given use-case. In the first phase, we introduce a zero-shot,
multi-stage reasoning approach which enables LLMs to produce and refine a label
taxonomy iteratively. In the second phase, LLMs are used as data labelers that
yield training samples so that lightweight supervised classifiers can be
reliably built, deployed, and served at scale. We apply TnT-LLM to the analysis
of user intent and conversational domain for Bing Copilot (formerly Bing Chat),
an open-domain chat-based search engine. Extensive experiments using both human
and automatic evaluation metrics demonstrate that TnT-LLM generates more
accurate and relevant label taxonomies when compared against state-of-the-art
baselines, and achieves a favorable balance between accuracy and efficiency for
classification at scale. We also share our practical experiences and insights
on the challenges and opportunities of using LLMs for large-scale text mining
in real-world applications. | cs.CL | TnT-LLM: Text Mining at Scale with Large Language Models |
2024-03-18T18:39:53Z | http://arxiv.org/pdf/2403.12171v1 | http://arxiv.org/abs/2403.12171v1 | Weikang Zhou, Xiao Wang, Limao Xiong, Han Xia, Yingshuang Gu, Mingxu Chai, Fukang Zhu, Caishuang Huang, Shihan Dou, Zhiheng Xi, Rui Zheng, Songyang Gao, Yicheng Zou, Hang Yan, Yifan Le, Ruohui Wang, Lijun Li, Jing Shao, Tao Gui, Qi Zhang, Xuanjing Huang | 2024-03-18T18:39:53Z | cs.CL, cs.AI | Jailbreak attacks are crucial for identifying and mitigating the security
vulnerabilities of Large Language Models (LLMs). They are designed to bypass
safeguards and elicit prohibited outputs. However, due to significant
differences among various jailbreak methods, there is no standard
implementation framework available for the community, which limits
comprehensive security evaluations. This paper introduces EasyJailbreak, a
unified framework simplifying the construction and evaluation of jailbreak
attacks against LLMs. It builds jailbreak attacks using four components:
Selector, Mutator, Constraint, and Evaluator. This modular framework enables
researchers to easily construct attacks from combinations of novel and existing
components. So far, EasyJailbreak supports 11 distinct jailbreak methods and
facilitates the security validation of a broad spectrum of LLMs. Our validation
across 10 distinct LLMs reveals a significant vulnerability, with an average
breach probability of 60% under various jailbreaking attacks. Notably, even
advanced models like GPT-3.5-Turbo and GPT-4 exhibit average Attack Success
Rates (ASR) of 57% and 33%, respectively. We have released a wealth of
resources for researchers, including a web platform, PyPI published package,
screencast video, and experimental outputs. | cs.CL | EasyJailbreak: A Unified Framework for Jailbreaking Large Language Models |
2024-03-18T18:08:44Z | http://arxiv.org/pdf/2403.12151v2 | http://arxiv.org/abs/2403.12151v2 | Filippos Gouidis, Katerina Papantoniou, Konstantinos Papoutsakis Theodore Patkos, Antonis Argyros, Dimitris Plexousakis | 2024-03-25T18:50:06Z | cs.AI, cs.CL, cs.CV, cs.LG | Domain-specific knowledge can significantly contribute to addressing a wide
variety of vision tasks. However, the generation of such knowledge entails
considerable human labor and time costs. This study investigates the potential
of Large Language Models (LLMs) in generating and providing domain-specific
information through semantic embeddings. To achieve this, an LLM is integrated
into a pipeline that utilizes Knowledge Graphs and pre-trained semantic vectors
in the context of the Vision-based Zero-shot Object State Classification task.
We thoroughly examine the behavior of the LLM through an extensive ablation
study. Our findings reveal that the integration of LLM-based embeddings, in
combination with general-purpose pre-trained embeddings, leads to substantial
performance improvements. Drawing insights from this ablation study, we conduct
a comparative analysis against competing models, thereby highlighting the
state-of-the-art performance achieved by the proposed approach. | cs.AI | Fusing Domain-Specific Content from Large Language Models into Knowledge Graphs for Enhanced Zero Shot Object State Classification |
2024-03-18T18:01:26Z | http://arxiv.org/pdf/2403.12145v1 | http://arxiv.org/abs/2403.12145v1 | Ashwin Daswani, Rohan Sawant, Najoung Kim | 2024-03-18T18:01:26Z | cs.CL | Sensitivity to false assumptions (or false premises) in information-seeking
questions is critical for robust question-answering (QA) systems. Recent work
has shown that false assumptions in naturally occurring questions pose
challenges to current models, with low performance on both generative QA and
simple detection tasks (Kim et al. 2023). However, the focus of existing work
on naturally occurring questions leads to a gap in the analysis of model
behavior on the long tail of the distribution of possible questions. To this
end, we introduce Syn-(QA)$^2$, a set of two synthetically generated QA
datasets: one generated using perturbed relations from Wikidata, and the other
by perturbing HotpotQA (Yang et al. 2018). Our findings from evaluating a range
of large language models are threefold: (1) false assumptions in QA are
challenging, echoing the findings of prior work, (2) the binary detection task
is challenging even compared to the difficulty of generative QA itself,
possibly due to the linguistic structure of the problem, and (3) the detection
task is more challenging with long-tail questions compared to naturally
occurring questions, highlighting the utility of our synthetic datasets and
generation method. | cs.CL | Syn-QA2: Evaluating False Assumptions in Long-tail Questions with Synthetic QA Datasets |
2024-03-18T17:57:09Z | http://arxiv.org/pdf/2403.12027v2 | http://arxiv.org/abs/2403.12027v2 | Kung-Hsiang Huang, Hou Pong Chan, Yi R. Fung, Haoyi Qiu, Mingyang Zhou, Shafiq Joty, Shih-Fu Chang, Heng Ji | 2024-03-25T17:39:10Z | cs.CL, cs.AI, cs.CV | Data visualization in the form of charts plays a pivotal role in data
analysis, offering critical insights and aiding in informed decision-making.
Automatic chart understanding has witnessed significant advancements with the
rise of large foundation models in recent years. Foundation models, such as
large language models, have revolutionized various natural language processing
tasks and are increasingly being applied to chart understanding tasks. This
survey paper provides a comprehensive overview of the recent developments,
challenges, and future directions in chart understanding within the context of
these foundation models. We review fundamental building blocks crucial for
studying chart understanding tasks. Additionally, we explore various tasks and
their evaluation metrics and sources of both charts and textual inputs. Various
modeling strategies are then examined, encompassing both classification-based
and generation-based approaches, along with tool augmentation techniques that
enhance chart understanding performance. Furthermore, we discuss the
state-of-the-art performance of each task and discuss how we can improve the
performance. Challenges and future directions are addressed, highlighting the
importance of several topics, such as domain-specific charts, lack of efforts
in developing evaluation metrics, and agent-oriented settings. This survey
paper serves as a comprehensive resource for researchers and practitioners in
the fields of natural language processing, computer vision, and data analysis,
providing valuable insights and directions for future research in chart
understanding leveraging large foundation models. The studies mentioned in this
paper, along with emerging new research, will be continually updated at:
https://github.com/khuangaf/Awesome-Chart-Understanding. | cs.CL | From Pixels to Insights: A Survey on Automatic Chart Understanding in the Era of Large Foundation Models |
2024-03-18T17:57:02Z | http://arxiv.org/pdf/2403.12026v1 | http://arxiv.org/abs/2403.12026v1 | Debidatta Dwibedi, Vidhi Jain, Jonathan Tompson, Andrew Zisserman, Yusuf Aytar | 2024-03-18T17:57:02Z | cs.CV, cs.AI, cs.CL, cs.LG | We introduce a versatile $\textit{flexible-captioning}$ vision-language model
(VLM) capable of generating region-specific descriptions of varying lengths.
The model, FlexCap, is trained to produce length-conditioned captions for input
bounding boxes, and this allows control over the information density of its
output, with descriptions ranging from concise object labels to detailed
captions. To achieve this we create large-scale training datasets of image
region descriptions of varying length, starting from captioned images. This
flexible-captioning capability has several valuable applications.
First, FlexCap demonstrates superior performance in dense captioning tasks on
the Visual Genome dataset. Second, a visual question answering (VQA) system can
be built by employing FlexCap to generate localized descriptions as inputs to a
large language model. The resulting system achieves state-of-the-art zero-shot
performance on a number of VQA datasets. We also demonstrate a
$\textit{localize-then-describe}$ approach with FlexCap can be better at
open-ended object detection than a $\textit{describe-then-localize}$ approach
with other VLMs. We highlight a novel characteristic of FlexCap, which is its
ability to extract diverse visual information through prefix conditioning.
Finally, we qualitatively demonstrate FlexCap's broad applicability in tasks
such as image labeling, object attribute recognition, and visual dialog.
Project webpage: https://flex-cap.github.io . | cs.CV | FlexCap: Generating Rich, Localized, and Flexible Captions in Images |
2024-03-18T17:56:37Z | http://arxiv.org/pdf/2403.12025v1 | http://arxiv.org/abs/2403.12025v1 | Stephen R. Pfohl, Heather Cole-Lewis, Rory Sayres, Darlene Neal, Mercy Asiedu, Awa Dieng, Nenad Tomasev, Qazi Mamunur Rashid, Shekoofeh Azizi, Negar Rostamzadeh, Liam G. McCoy, Leo Anthony Celi, Yun Liu, Mike Schaekermann, Alanna Walton, Alicia Parrish, Chirag Nagpal, Preeti Singh, Akeiylah Dewitt, Philip Mansfield, Sushant Prakash, Katherine Heller, Alan Karthikesalingam, Christopher Semturs, Joelle Barral, Greg Corrado, Yossi Matias, Jamila Smith-Loud, Ivor Horn, Karan Singhal | 2024-03-18T17:56:37Z | cs.CY, cs.CL, cs.LG | Large language models (LLMs) hold immense promise to serve complex health
information needs but also have the potential to introduce harm and exacerbate
health disparities. Reliably evaluating equity-related model failures is a
critical step toward developing systems that promote health equity. In this
work, we present resources and methodologies for surfacing biases with
potential to precipitate equity-related harms in long-form, LLM-generated
answers to medical questions and then conduct an empirical case study with
Med-PaLM 2, resulting in the largest human evaluation study in this area to
date. Our contributions include a multifactorial framework for human assessment
of LLM-generated answers for biases, and EquityMedQA, a collection of seven
newly-released datasets comprising both manually-curated and LLM-generated
questions enriched for adversarial queries. Both our human assessment framework
and dataset design process are grounded in an iterative participatory approach
and review of possible biases in Med-PaLM 2 answers to adversarial queries.
Through our empirical study, we find that the use of a collection of datasets
curated through a variety of methodologies, coupled with a thorough evaluation
protocol that leverages multiple assessment rubric designs and diverse rater
groups, surfaces biases that may be missed via narrower evaluation approaches.
Our experience underscores the importance of using diverse assessment
methodologies and involving raters of varying backgrounds and expertise. We
emphasize that while our framework can identify specific forms of bias, it is
not sufficient to holistically assess whether the deployment of an AI system
promotes equitable health outcomes. We hope the broader community leverages and
builds on these tools and methods towards realizing a shared goal of LLMs that
promote accessible and equitable healthcare for all. | cs.CY | A Toolbox for Surfacing Health Equity Harms and Biases in Large Language Models |
2024-03-18T17:56:13Z | http://arxiv.org/pdf/2403.12024v2 | http://arxiv.org/abs/2403.12024v2 | Bo-Han Lu, Yi-Hsuan Lin, En-Shiun Annie Lee, Richard Tzong-Han Tsai | 2024-05-14T05:49:00Z | cs.CL | Machine translation focuses mainly on high-resource languages (HRLs), while
low-resource languages (LRLs) like Taiwanese Hokkien are relatively
under-explored. The study aims to address this gap by developing a dual
translation model between Taiwanese Hokkien and both Traditional Mandarin
Chinese and English. We employ a pre-trained LLaMA 2-7B model specialized in
Traditional Mandarin Chinese to leverage the orthographic similarities between
Taiwanese Hokkien Han and Traditional Mandarin Chinese. Our comprehensive
experiments involve translation tasks across various writing systems of
Taiwanese Hokkien as well as between Taiwanese Hokkien and other HRLs. We find
that the use of a limited monolingual corpus still further improves the model's
Taiwanese Hokkien capabilities. We then utilize our translation model to
standardize all Taiwanese Hokkien writing systems into Hokkien Han, resulting
in further performance improvements. Additionally, we introduce an evaluation
method incorporating back-translation and GPT-4 to ensure reliable translation
quality assessment even for LRLs. The study contributes to narrowing the
resource gap for Taiwanese Hokkien and empirically investigates the advantages
and limitations of pre-training and fine-tuning based on LLaMA 2. | cs.CL | Enhancing Taiwanese Hokkien Dual Translation by Exploring and Standardizing of Four Writing Systems |
2024-03-18T17:52:57Z | http://arxiv.org/pdf/2403.12017v1 | http://arxiv.org/abs/2403.12017v1 | Hao Sun | 2024-03-18T17:52:57Z | cs.LG, cs.AI, cs.CL | The prevailing approach to aligning Large Language Models (LLMs) typically
relies on human or AI feedback and assumes access to specific types of
preference datasets. In our work, we question the efficacy of such datasets and
explore various scenarios where alignment with expert demonstrations proves
more realistic. We build a sequential decision-making framework to formulate
the problem of aligning LLMs using demonstration datasets. Drawing insights
from inverse reinforcement learning and imitation learning, we introduce
various approaches for divergence minimization in the LLM alignment tasks. Our
analysis highlights the mass-covering and mode-seeking behaviors of these
different approaches. Inclusively, we examine the pros and cons of the
classical supervised fine-tuning method, elaborating on scenarios where
different methods shine. | cs.LG | Supervised Fine-Tuning as Inverse Reinforcement Learning |
2024-03-18T17:51:16Z | http://arxiv.org/pdf/2403.12014v1 | http://arxiv.org/abs/2403.12014v1 | Abhay Zala, Jaemin Cho, Han Lin, Jaehong Yoon, Mohit Bansal | 2024-03-18T17:51:16Z | cs.CL, cs.AI, cs.LG | Recent SOTA approaches for embodied learning via interaction directly employ
large language models (LLMs) as agents to determine the next steps in an
environment. Due to their world knowledge and reasoning capabilities, LLM
agents achieve stronger performance than previous smaller agents based on
reinforcement learning (RL); however, frequently calling LLMs is slow and
expensive. Instead of directly employing LLMs as agents, can we use LLMs'
reasoning capabilities to adaptively create training environments to help
smaller embodied RL agents learn useful skills that they are weak at? We
propose EnvGen, a novel framework to address this question. First, we prompt an
LLM to generate training environments that allow agents to quickly learn
different tasks in parallel. Concretely, the LLM is given the task description
and simulator objectives that the agents should learn and is then asked to
generate a set of environment configurations (e.g., different terrains, items
given to agents, etc.). Next, we train a small RL agent in a mixture of the
original and LLM-generated environments. Then, we enable the LLM to
continuously adapt the generated environments to progressively improve the
skills that the agent is weak at, by providing feedback to the LLM in the form
of the agent's performance. We demonstrate the usefulness of EnvGen with
comprehensive experiments in Crafter and Heist environments. We find that a
small RL agent trained with EnvGen can outperform SOTA methods, including a
GPT-4 agent, and learns long-horizon tasks significantly faster. We show
qualitatively how the LLM adapts training environments to help improve RL
agents' weaker skills over time. Additionally, EnvGen is substantially more
efficient as it only uses a small number of LLM calls (e.g., 4 in total),
whereas LLM agents require thousands of LLM calls. Lastly, we present detailed
ablation studies for our design choices. | cs.CL | EnvGen: Generating and Adapting Environments via LLMs for Training Embodied Agents |
2024-03-18T17:32:32Z | http://arxiv.org/pdf/2403.12766v2 | http://arxiv.org/abs/2403.12766v2 | Cunxiang Wang, Ruoxi Ning, Boqi Pan, Tonghui Wu, Qipeng Guo, Cheng Deng, Guangsheng Bao, Xiangkun Hu, Zheng Zhang, Qian Wang, Yue Zhang | 2024-06-17T13:53:15Z | cs.CL | The rapid advancement of Large Language Models (LLMs) has introduced a new
frontier in natural language processing, particularly in understanding and
processing long-context information. However, the evaluation of these models'
long-context abilities remains a challenge due to the limitations of current
benchmarks. To address this gap, we introduce NovelQA, a benchmark specifically
designed to test the capabilities of LLMs with extended texts. Constructed from
English novels, NovelQA offers a unique blend of complexity, length, and
narrative coherence, making it an ideal tool for assessing deep textual
understanding in LLMs. This paper presents the design and construction of
NovelQA, highlighting its manual annotation, and diverse question types. Our
evaluation of Long-context LLMs on NovelQA reveals significant insights into
the models' performance, particularly emphasizing the challenges they face with
multi-hop reasoning, detail-oriented questions, and extremely long input with
an average length more than 200,000 tokens. The results underscore the
necessity for further advancements in LLMs to improve their long-context
comprehension. | cs.CL | NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.