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
|
|---|---|---|---|---|---|---|---|---|
2024-05-07T01:40:23Z | http://arxiv.org/pdf/2405.03932v2 | http://arxiv.org/abs/2405.03932v2 | Tyler Bikaun, Michael Stewart, Wei Liu | 2024-05-08T00:18:45Z | cs.AI, cs.CL | This paper presents CleanGraph, an interactive web-based tool designed to
facilitate the refinement and completion of knowledge graphs. Maintaining the
reliability of knowledge graphs, which are grounded in high-quality and
error-free facts, is crucial for real-world applications such as
question-answering and information retrieval systems. These graphs are often
automatically assembled from textual sources by extracting semantic triples via
information extraction. However, assuring the quality of these extracted
triples, especially when dealing with large or low-quality datasets, can pose a
significant challenge and adversely affect the performance of downstream
applications. CleanGraph allows users to perform Create, Read, Update, and
Delete (CRUD) operations on their graphs, as well as apply models in the form
of plugins for graph refinement and completion tasks. These functionalities
enable users to enhance the integrity and reliability of their graph data. A
demonstration of CleanGraph and its source code can be accessed at
https://github.com/nlp-tlp/CleanGraph under the MIT License. | cs.AI | CleanGraph: Human-in-the-loop Knowledge Graph Refinement and Completion |
2024-05-07T00:38:34Z | http://arxiv.org/pdf/2405.03920v1 | http://arxiv.org/abs/2405.03920v1 | Dainis Boumber, Rakesh M. Verma, Fatima Zahra Qachfar | 2024-05-07T00:38:34Z | cs.CL, cs.AI, cs.MM, I.2.6; I.2.7; I.2.10; K.4.4 | Deception, a prevalent aspect of human communication, has undergone a
significant transformation in the digital age. With the globalization of online
interactions, individuals are communicating in multiple languages and mixing
languages on social media, with varied data becoming available in each language
and dialect. At the same time, the techniques for detecting deception are
similar across the board. Recent studies have shown the possibility of the
existence of universal linguistic cues to deception across domains within the
English language; however, the existence of such cues in other languages
remains unknown. Furthermore, the practical task of deception detection in
low-resource languages is not a well-studied problem due to the lack of labeled
data. Another dimension of deception is multimodality. For example, a picture
with an altered caption in fake news or disinformation may exist. This paper
calls for a comprehensive investigation into the complexities of deceptive
language across linguistic boundaries and modalities within the realm of
computer security and natural language processing and the possibility of using
multilingual transformer models and labeled data in various languages to
universally address the task of deception detection. | cs.CL | A Roadmap for Multilingual, Multimodal Domain Independent Deception Detection |
2024-05-06T21:20:35Z | http://arxiv.org/pdf/2405.03862v1 | http://arxiv.org/abs/2405.03862v1 | Razan Baltaji, Babak Hemmatian, Lav R. Varshney | 2024-05-06T21:20:35Z | cs.AI, cs.CL, I.2.7 | This study explores the sources of instability in maintaining cultural
personas and opinions within multi-agent LLM systems. Drawing on simulations of
inter-cultural collaboration and debate, we analyze agents' pre- and
post-discussion private responses alongside chat transcripts to assess the
stability of cultural personas and the impact of opinion diversity on group
outcomes. Our findings suggest that multi-agent discussions can encourage
collective decisions that reflect diverse perspectives, yet this benefit is
tempered by the agents' susceptibility to conformity due to perceived peer
pressure and challenges in maintaining consistent personas and opinions.
Counterintuitively, instructions that encourage debate in support of one's
opinions increase the rate of inconstancy. Without addressing the factors we
identify, the full potential of multi-agent frameworks for producing more
culturally diverse AI outputs will remain untapped. | cs.AI | Conformity, Confabulation, and Impersonation: Persona Inconstancy in Multi-Agent LLM Collaboration |
2024-05-06T20:50:17Z | http://arxiv.org/pdf/2405.03845v1 | http://arxiv.org/abs/2405.03845v1 | Guy Azov, Tatiana Pelc, Adi Fledel Alon, Gila Kamhi | 2024-05-06T20:50:17Z | cs.CL, cs.AI | Previous studies have demonstrated that proactive interaction with user
reviews has a positive impact on the perception of app users and encourages
them to submit revised ratings. Nevertheless, developers encounter challenges
in managing a high volume of reviews, particularly in the case of popular apps
with a substantial influx of daily reviews. Consequently, there is a demand for
automated solutions aimed at streamlining the process of responding to user
reviews. To address this, we have developed a new system for generating
automatic responses by leveraging user-contributed documents with the help of
retrieval-augmented generation (RAG) and advanced Large Language Models (LLMs).
Our solution, named SCRABLE, represents an adaptive customer review response
automation that enhances itself with self-optimizing prompts and a judging
mechanism based on LLMs. Additionally, we introduce an automatic scoring
mechanism that mimics the role of a human evaluator to assess the quality of
responses generated in customer review domains. Extensive experiments and
analyses conducted on real-world datasets reveal that our method is effective
in producing high-quality responses, yielding improvement of more than 8.5%
compared to the baseline. Further validation through manual examination of the
generated responses underscores the efficacy our proposed system. | cs.CL | Self-Improving Customer Review Response Generation Based on LLMs |
2024-05-06T20:30:14Z | http://arxiv.org/pdf/2405.03832v3 | http://arxiv.org/abs/2405.03832v3 | Christopher Clarke, Roland Daynauth, Charlene Wilkinson, Hubert Devonish, Jason Mars | 2024-07-02T21:23:32Z | cs.CL, cs.AI | While major languages often enjoy substantial attention and resources, the
linguistic diversity across the globe encompasses a multitude of smaller,
indigenous, and regional languages that lack the same level of computational
support. One such region is the Caribbean. While commonly labeled as "English
speaking", the ex-British Caribbean region consists of a myriad of Creole
languages thriving alongside English. In this paper, we present Guylingo: a
comprehensive corpus designed for advancing NLP research in the domain of
Creolese (Guyanese English-lexicon Creole), the most widely spoken language in
the culturally rich nation of Guyana. We first outline our framework for
gathering and digitizing this diverse corpus, inclusive of colloquial
expressions, idioms, and regional variations in a low-resource language. We
then demonstrate the challenges of training and evaluating NLP models for
machine translation in Creole. Lastly, we discuss the unique opportunities
presented by recent NLP advancements for accelerating the formal adoption of
Creole languages as official languages in the Caribbean. | cs.CL | Guylingo: The Republic of Guyana Creole Corpora |
2024-05-06T19:00:31Z | http://arxiv.org/pdf/2405.03794v1 | http://arxiv.org/abs/2405.03794v1 | Dengyi Liu, Minghao Wang, Andrew G. Catlin | 2024-05-06T19:00:31Z | cs.CL | Academic researchers and social media entities grappling with the
identification of hate speech face significant challenges, primarily due to the
vast scale of data and the dynamic nature of hate speech. Given the ethical and
practical limitations of large predictive models like ChatGPT in directly
addressing such sensitive issues, our research has explored alternative
advanced transformer-based and generative AI technologies since 2019.
Specifically, we developed a new data labeling technique and established a
proof of concept targeting anti-Semitic hate speech, utilizing a variety of
transformer models such as BERT (arXiv:1810.04805), DistillBERT
(arXiv:1910.01108), RoBERTa (arXiv:1907.11692), and LLaMA-2 (arXiv:2307.09288),
complemented by the LoRA fine-tuning approach (arXiv:2106.09685). This paper
delineates and evaluates the comparative efficacy of these cutting-edge methods
in tackling the intricacies of hate speech detection, highlighting the need for
responsible and carefully managed AI applications within sensitive contexts. | cs.CL | Detecting Anti-Semitic Hate Speech using Transformer-based Large Language Models |
2024-05-06T18:45:18Z | http://arxiv.org/pdf/2405.03789v1 | http://arxiv.org/abs/2405.03789v1 | Korn Sooksatra, Bikram Khanal, Pablo Rivas | 2024-05-06T18:45:18Z | cs.LG, cs.AI, cs.CL, cs.CR, 68T01, 68T50, I.2.7 | Recently, with the advancement of deep learning, several applications in text
classification have advanced significantly. However, this improvement comes
with a cost because deep learning is vulnerable to adversarial examples. This
weakness indicates that deep learning is not very robust. Fortunately, the
input of a text classifier is discrete. Hence, it can prevent the classifier
from state-of-the-art attacks. Nonetheless, previous works have generated
black-box attacks that successfully manipulate the discrete values of the input
to find adversarial examples. Therefore, instead of changing the discrete
values, we transform the input into its embedding vector containing real values
to perform the state-of-the-art white-box attacks. Then, we convert the
perturbed embedding vector back into a text and name it an adversarial example.
In summary, we create a framework that measures the robustness of a text
classifier by using the gradients of the classifier. | cs.LG | On Adversarial Examples for Text Classification by Perturbing Latent Representations |
2024-05-06T18:09:32Z | http://arxiv.org/pdf/2405.06687v1 | http://arxiv.org/abs/2405.06687v1 | Damin Zhang, Yi Zhang, Geetanjali Bihani, Julia Rayz | 2024-05-06T18:09:32Z | cs.CL | With the impressive performance in various downstream tasks, large language
models (LLMs) have been widely integrated into production pipelines, like
recruitment and recommendation systems. A known issue of models trained on
natural language data is the presence of human biases, which can impact the
fairness of the system. This paper investigates LLMs' behavior with respect to
gender stereotypes, in the context of occupation decision making. Our framework
is designed to investigate and quantify the presence of gender stereotypes in
LLMs' behavior via multi-round question answering. Inspired by prior works, we
construct a dataset by leveraging a standard occupation classification
knowledge base released by authoritative agencies. We tested three LLMs
(RoBERTa-large, GPT-3.5-turbo, and Llama2-70b-chat) and found that all models
exhibit gender stereotypes analogous to human biases, but with different
preferences. The distinct preferences of GPT-3.5-turbo and Llama2-70b-chat may
imply the current alignment methods are insufficient for debiasing and could
introduce new biases contradicting the traditional gender stereotypes. | cs.CL | Hire Me or Not? Examining Language Model's Behavior with Occupation Attributes |
2024-05-06T18:02:00Z | http://arxiv.org/pdf/2405.03764v1 | http://arxiv.org/abs/2405.03764v1 | Wenjie Zhou, Zhenxin Ding, Xiaodong Zhang, Haibo Shi, Junfeng Wang, Dawei Yin | 2024-05-06T18:02:00Z | cs.CL, cs.IR | Pre-trained language models have become an integral component of
question-answering systems, achieving remarkable performance. For practical
deployment, it is critical to carry out knowledge distillation to preserve high
performance under computational constraints. In this paper, we address a key
question: given the importance of unsupervised distillation for student
performance, how does one effectively ensemble knowledge from multiple teachers
at this stage without the guidance of ground-truth labels? We propose a novel
algorithm, GOVERN, to tackle this issue. GOVERN has demonstrated significant
improvements in both offline and online experiments. The proposed algorithm has
been successfully deployed in a real-world commercial question-answering
system. | cs.CL | GOVERN: Gradient Orientation Vote Ensemble for Multi-Teacher Reinforced Distillation |
2024-05-06T17:59:36Z | http://arxiv.org/pdf/2405.03689v1 | http://arxiv.org/abs/2405.03689v1 | Sanjay Subramanian, Evonne Ng, Lea Müller, Dan Klein, Shiry Ginosar, Trevor Darrell | 2024-05-06T17:59:36Z | cs.CV, cs.CL | We present a zero-shot pose optimization method that enforces accurate
physical contact constraints when estimating the 3D pose of humans. Our central
insight is that since language is often used to describe physical interaction,
large pretrained text-based models can act as priors on pose estimation.
We can thus leverage this insight to improve pose estimation by converting
natural language descriptors, generated by a large multimodal model (LMM), into
tractable losses to constrain the 3D pose optimization. Despite its simplicity,
our method produces surprisingly compelling pose reconstructions of people in
close contact, correctly capturing the semantics of the social and physical
interactions. We demonstrate that our method rivals more complex
state-of-the-art approaches that require expensive human annotation of contact
points and training specialized models. Moreover, unlike previous approaches,
our method provides a unified framework for resolving self-contact and
person-to-person contact. | cs.CV | Pose Priors from Language Models |
2024-05-06T17:59:07Z | http://arxiv.org/pdf/2405.03688v1 | http://arxiv.org/abs/2405.03688v1 | Keith Burghardt, Kai Chen, Kristina Lerman | 2024-05-06T17:59:07Z | cs.CL, cs.LG | Adversarial information operations can destabilize societies by undermining
fair elections, manipulating public opinions on policies, and promoting scams.
Despite their widespread occurrence and potential impacts, our understanding of
influence campaigns is limited by manual analysis of messages and subjective
interpretation of their observable behavior. In this paper, we explore whether
these limitations can be mitigated with large language models (LLMs), using
GPT-3.5 as a case-study for coordinated campaign annotation. We first use
GPT-3.5 to scrutinize 126 identified information operations spanning over a
decade. We utilize a number of metrics to quantify the close (if imperfect)
agreement between LLM and ground truth descriptions. We next extract
coordinated campaigns from two large multilingual datasets from X (formerly
Twitter) that respectively discuss the 2022 French election and 2023 Balikaran
Philippine-U.S. military exercise in 2023. For each coordinated campaign, we
use GPT-3.5 to analyze posts related to a specific concern and extract goals,
tactics, and narrative frames, both before and after critical events (such as
the date of an election). While the GPT-3.5 sometimes disagrees with subjective
interpretation, its ability to summarize and interpret demonstrates LLMs'
potential to extract higher-order indicators from text to provide a more
complete picture of the information campaigns compared to previous methods. | cs.CL | Large Language Models Reveal Information Operation Goals, Tactics, and Narrative Frames |
2024-05-06T17:57:27Z | http://arxiv.org/pdf/2405.03685v1 | http://arxiv.org/abs/2405.03685v1 | Jang Hyun Cho, Boris Ivanovic, Yulong Cao, Edward Schmerling, Yue Wang, Xinshuo Weng, Boyi Li, Yurong You, Philipp Krähenbühl, Yan Wang, Marco Pavone | 2024-05-06T17:57:27Z | cs.CV, cs.AI, cs.CL, cs.LG | Multi-modal large language models (MLLMs) have shown incredible capabilities
in a variety of 2D vision and language tasks. We extend MLLMs' perceptual
capabilities to ground and reason about images in 3-dimensional space. To that
end, we first develop a large-scale pre-training dataset for 2D and 3D called
LV3D by combining multiple existing 2D and 3D recognition datasets under a
common task formulation: as multi-turn question-answering. Next, we introduce a
new MLLM named Cube-LLM and pre-train it on LV3D. We show that pure data
scaling makes a strong 3D perception capability without 3D specific
architectural design or training objective. Cube-LLM exhibits intriguing
properties similar to LLMs: (1) Cube-LLM can apply chain-of-thought prompting
to improve 3D understanding from 2D context information. (2) Cube-LLM can
follow complex and diverse instructions and adapt to versatile input and output
formats. (3) Cube-LLM can be visually prompted such as 2D box or a set of
candidate 3D boxes from specialists. Our experiments on outdoor benchmarks
demonstrate that Cube-LLM significantly outperforms existing baselines by 21.3
points of AP-BEV on the Talk2Car dataset for 3D grounded reasoning and 17.7
points on the DriveLM dataset for complex reasoning about driving scenarios,
respectively. Cube-LLM also shows competitive results in general MLLM
benchmarks such as refCOCO for 2D grounding with (87.0) average score, as well
as visual question answering benchmarks such as VQAv2, GQA, SQA, POPE, etc. for
complex reasoning. Our project is available at
https://janghyuncho.github.io/Cube-LLM. | cs.CV | Language-Image Models with 3D Understanding |
2024-05-06T17:53:33Z | http://arxiv.org/pdf/2405.03677v1 | http://arxiv.org/abs/2405.03677v1 | Clayton Cohn, Caitlin Snyder, Justin Montenegro, Gautam Biswas | 2024-05-06T17:53:33Z | cs.CL | LLMs have demonstrated proficiency in contextualizing their outputs using
human input, often matching or beating human-level performance on a variety of
tasks. However, LLMs have not yet been used to characterize synergistic
learning in students' collaborative discourse. In this exploratory work, we
take a first step towards adopting a human-in-the-loop prompt engineering
approach with GPT-4-Turbo to summarize and categorize students' synergistic
learning during collaborative discourse. Our preliminary findings suggest
GPT-4-Turbo may be able to characterize students' synergistic learning in a
manner comparable to humans and that our approach warrants further
investigation. | cs.CL | Towards A Human-in-the-Loop LLM Approach to Collaborative Discourse Analysis |
2024-05-06T17:41:33Z | http://arxiv.org/pdf/2405.15793v2 | http://arxiv.org/abs/2405.15793v2 | John Yang, Carlos E. Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, Ofir Press | 2024-05-30T19:09:01Z | cs.SE, cs.AI, cs.CL, cs.HC, cs.LG | Language model (LM) agents are increasingly being used to automate
complicated tasks in digital environments. Just as humans benefit from powerful
software applications, such as integrated development environments, for complex
tasks like software engineering, we posit that LM agents represent a new
category of end users with their own needs and abilities, and would benefit
from specially-built interfaces to the software they use. We investigate how
interface design affects the performance of language model agents. As a result
of this exploration, we introduce SWE-agent: a system that facilitates LM
agents to autonomously use computers to solve software engineering tasks.
SWE-agent's custom agent-computer interface (ACI) significantly enhances an
agent's ability to create and edit code files, navigate entire repositories,
and execute tests and other programs. We evaluate SWE-agent on SWE-bench and
HumanEvalFix, achieving state-of-the-art performance on both with a pass@1 rate
of 12.5% and 87.7%, respectively, far exceeding the previous state-of-the-art
achieved with non-interactive LMs. Finally, we provide insight on how the
design of the ACI can impact agents' behavior and performance. | cs.SE | SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering |
2024-05-06T16:04:03Z | http://arxiv.org/pdf/2405.03595v1 | http://arxiv.org/abs/2405.03595v1 | Sophie Ostmeier, Justin Xu, Zhihong Chen, Maya Varma, Louis Blankemeier, Christian Bluethgen, Arne Edward Michalson, Michael Moseley, Curtis Langlotz, Akshay S Chaudhari, Jean-Benoit Delbrouck | 2024-05-06T16:04:03Z | cs.CL, cs.AI | Evaluating radiology reports is a challenging problem as factual correctness
is extremely important due to the need for accurate medical communication about
medical images. Existing automatic evaluation metrics either suffer from
failing to consider factual correctness (e.g., BLEU and ROUGE) or are limited
in their interpretability (e.g., F1CheXpert and F1RadGraph). In this paper, we
introduce GREEN (Generative Radiology Report Evaluation and Error Notation), a
radiology report generation metric that leverages the natural language
understanding of language models to identify and explain clinically significant
errors in candidate reports, both quantitatively and qualitatively. Compared to
current metrics, GREEN offers: 1) a score aligned with expert preferences, 2)
human interpretable explanations of clinically significant errors, enabling
feedback loops with end-users, and 3) a lightweight open-source method that
reaches the performance of commercial counterparts. We validate our GREEN
metric by comparing it to GPT-4, as well as to error counts of 6 experts and
preferences of 2 experts. Our method demonstrates not only higher correlation
with expert error counts, but simultaneously higher alignment with expert
preferences when compared to previous approaches." | cs.CL | GREEN: Generative Radiology Report Evaluation and Error Notation |
2024-05-06T16:03:32Z | http://arxiv.org/pdf/2405.03594v1 | http://arxiv.org/abs/2405.03594v1 | Abhinav Agarwalla, Abhay Gupta, Alexandre Marques, Shubhra Pandit, Michael Goin, Eldar Kurtic, Kevin Leong, Tuan Nguyen, Mahmoud Salem, Dan Alistarh, Sean Lie, Mark Kurtz | 2024-05-06T16:03:32Z | cs.CL, cs.AI | Large language models (LLMs) have revolutionized Natural Language Processing
(NLP), but their size creates computational bottlenecks. We introduce a novel
approach to create accurate, sparse foundational versions of performant LLMs
that achieve full accuracy recovery for fine-tuning tasks at up to 70%
sparsity. We achieve this for the LLaMA-2 7B model by combining the SparseGPT
one-shot pruning method and sparse pretraining of those models on a subset of
the SlimPajama dataset mixed with a Python subset of The Stack dataset. We
exhibit training acceleration due to sparsity on Cerebras CS-3 chips that
closely matches theoretical scaling. In addition, we establish inference
acceleration of up to 3x on CPUs by utilizing Neural Magic's DeepSparse engine
and 1.7x on GPUs through Neural Magic's nm-vllm engine. The above gains are
realized via sparsity alone, thus enabling further gains through additional use
of quantization. Specifically, we show a total speedup on CPUs for
sparse-quantized LLaMA models of up to 8.6x. We demonstrate these results
across diverse, challenging tasks, including chat, instruction following, code
generation, arithmetic reasoning, and summarization to prove their generality.
This work paves the way for rapidly creating smaller and faster LLMs without
sacrificing accuracy. | cs.CL | Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment |
2024-05-06T15:20:30Z | http://arxiv.org/pdf/2405.03553v2 | http://arxiv.org/abs/2405.03553v2 | Guoxin Chen, Minpeng Liao, Chengxi Li, Kai Fan | 2024-05-23T05:07:24Z | cs.CL, cs.AI | Recent advancements in large language models (LLMs) have substantially
enhanced their mathematical reasoning abilities. However, these models still
struggle with complex problems that require multiple reasoning steps,
frequently leading to logical or numerical errors. While numerical mistakes can
be largely addressed by integrating a code interpreter, identifying logical
errors within intermediate steps is more challenging. Moreover, manually
annotating these steps for training is not only expensive but also
labor-intensive, requiring the expertise of professional annotators. In our
study, we introduce an innovative approach that bypasses the need for process
annotations (from human or GPTs) by utilizing the Monte Carlo Tree Search
(MCTS) framework. This technique automatically generates both the process
supervision and the step-level evaluation signals. Our method iteratively
trains the policy and value models, leveraging the capabilities of a
well-pretrained LLM to progressively enhance its mathematical reasoning skills.
Furthermore, we propose an efficient inference strategy-step-level beam search,
where the value model is crafted to assist the policy model (i.e., LLM) in
navigating more effective reasoning paths, rather than solely relying on prior
probabilities. The experimental results on both in-domain and out-of-domain
datasets demonstrate that even without GPT-4 or human-annotated process
supervision, our AlphaMath framework achieves comparable or superior results to
previous state-of-the-art methods. | cs.CL | AlphaMath Almost Zero: process Supervision without process |
2024-05-06T15:11:38Z | http://arxiv.org/pdf/2405.03548v4 | http://arxiv.org/abs/2405.03548v4 | Xiang Yue, Tuney Zheng, Ge Zhang, Wenhu Chen | 2024-05-23T16:34:35Z | cs.CL | Instruction tuning improves the reasoning abilities of large language models
(LLMs), with data quality and scalability being the crucial factors. Most
instruction tuning data come from human crowd-sourcing or GPT-4 distillation.
We propose a paradigm to efficiently harvest 10 million naturally existing
instruction data from the pre-training web corpus to enhance LLM reasoning. Our
approach involves (1) recalling relevant documents, (2) extracting
instruction-response pairs, and (3) refining the extracted pairs using
open-source LLMs. Fine-tuning base LLMs on this dataset, we build MAmmoTH2
models, which significantly boost performance on reasoning benchmarks. Notably,
MAmmoTH2-7B's (Mistral) performance increases from 11% to 36.7% on MATH and
from 36% to 68.4% on GSM8K without training on any in-domain data. Further
training MAmmoTH2 on public instruction tuning datasets yields MAmmoTH2-Plus,
achieving state-of-the-art performance on several reasoning and chatbot
benchmarks. Our work demonstrates how to harvest large-scale, high-quality
instruction data without costly human annotation or GPT-4 distillation,
providing a new paradigm for building better instruction tuning data. | cs.CL | MAmmoTH2: Scaling Instructions from the Web |
2024-05-06T14:21:52Z | http://arxiv.org/pdf/2405.06686v1 | http://arxiv.org/abs/2405.06686v1 | Muhammad U. Nasir, Steven James, Julian Togelius | 2024-05-06T14:21:52Z | cs.CL, cs.AI | Large Language Models (LLMs) have proven their worth across a diverse
spectrum of disciplines. LLMs have shown great potential in Procedural Content
Generation (PCG) as well, but directly generating a level through a pre-trained
LLM is still challenging. This work introduces Word2World, a system that
enables LLMs to procedurally design playable games through stories, without any
task-specific fine-tuning. Word2World leverages the abilities of LLMs to create
diverse content and extract information. Combining these abilities, LLMs can
create a story for the game, design narrative, and place tiles in appropriate
places to create coherent worlds and playable games. We test Word2World with
different LLMs and perform a thorough ablation study to validate each step. We
open-source the code at https://github.com/umair-nasir14/Word2World. | cs.CL | Word2World: Generating Stories and Worlds through Large Language Models |
2024-05-06T13:23:57Z | http://arxiv.org/pdf/2405.03452v2 | http://arxiv.org/abs/2405.03452v2 | Jairo Gudiño-Rosero, Umberto Grandi, César A. Hidalgo | 2024-05-07T08:57:18Z | cs.CY, cs.AI, cs.CL | We explore the capabilities of an augmented democracy system built on
off-the-shelf LLMs fine-tuned on data summarizing individual preferences across
67 policy proposals collected during the 2022 Brazilian presidential elections.
We use a train-test cross-validation setup to estimate the accuracy with which
the LLMs predict both: a subject's individual political choices and the
aggregate preferences of the full sample of participants. At the individual
level, the accuracy of the out of sample predictions lie in the range 69%-76%
and are significantly better at predicting the preferences of liberal and
college educated participants. At the population level, we aggregate
preferences using an adaptation of the Borda score and compare the ranking of
policy proposals obtained from a probabilistic sample of participants and from
data augmented using LLMs. We find that the augmented data predicts the
preferences of the full population of participants better than probabilistic
samples alone when these represent less than 30% to 40% of the total
population. These results indicate that LLMs are potentially useful for the
construction of systems of augmented democracy. | cs.CY | Large Language Models (LLMs) as Agents for Augmented Democracy |
2024-05-06T12:54:41Z | http://arxiv.org/pdf/2405.06685v1 | http://arxiv.org/abs/2405.06685v1 | Edirlei Soares de Lima, Margot M. E. Neggers, Antonio L. Furtado | 2024-05-06T12:54:41Z | cs.CL | This paper shows how to construct genre patterns, whose purpose is to guide
interactive story composition in a way that enforces thematic consistency. To
start the discussion we argue, based on previous seminal works, for the
existence of five fundamental genres, namely comedy, romance - in the sense of
epic plots, flourishing since the twelfth century -, tragedy, satire, and
mystery. To construct the patterns, a simple two-phase process is employed:
first retrieving examples that match our genre characterizations, and then
applying a form of most specific generalization to the groups of examples in
order to find their commonalities. In both phases, AI agents are instrumental,
with our PatternTeller prototype being called to operate the story composition
process, offering the opportunity to generate stories from a given premise of
the user, to be developed under the guidance of the chosen pattern and trying
to accommodate the user's suggestions along the composition stages. | cs.CL | Multigenre AI-powered Story Composition |
2024-05-06T12:44:37Z | http://arxiv.org/pdf/2405.03425v1 | http://arxiv.org/abs/2405.03425v1 | Emre Onal, Klemens Flöge, Emma Caldwell, Arsen Sheverdin, Vincent Fortuin | 2024-05-06T12:44:37Z | cs.CL | Fine-tuned Large Language Models (LLMs) often suffer from overconfidence and
poor calibration, particularly when fine-tuned on small datasets. To address
these challenges, we propose a simple combination of Low-Rank Adaptation (LoRA)
with Gaussian Stochastic Weight Averaging (SWAG), facilitating approximate
Bayesian inference in LLMs. Through extensive testing across several Natural
Language Processing (NLP) benchmarks, we demonstrate that our straightforward
and computationally efficient approach improves model generalization and
calibration. We further show that our method exhibits greater robustness
against distribution shift, as reflected in its performance on
out-of-distribution tasks. | cs.CL | Gaussian Stochastic Weight Averaging for Bayesian Low-Rank Adaptation of Large Language Models |
2024-05-06T11:45:59Z | http://arxiv.org/pdf/2405.03387v1 | http://arxiv.org/abs/2405.03387v1 | Hang Yuan, Zhongyue Che, Shao Li, Yue Zhang, Xiaomeng Hu, Siyang Luo | 2024-05-06T11:45:59Z | cs.CL | Given the rapid advancement of large-scale language models, artificial
intelligence (AI) models, like ChatGPT, are playing an increasingly prominent
role in human society. However, to ensure that artificial intelligence models
benefit human society, we must first fully understand the similarities and
differences between the human-like characteristics exhibited by artificial
intelligence models and real humans, as well as the cultural stereotypes and
biases that artificial intelligence models may exhibit in the process of
interacting with humans. This study first measured ChatGPT in 84 dimensions of
psychological characteristics, revealing differences between ChatGPT and human
norms in most dimensions as well as in high-dimensional psychological
representations. Additionally, through the measurement of ChatGPT in 13
dimensions of cultural values, it was revealed that ChatGPT's cultural value
patterns are dissimilar to those of various countries/regions worldwide.
Finally, an analysis of ChatGPT's performance in eight decision-making tasks
involving interactions with humans from different countries/regions revealed
that ChatGPT exhibits clear cultural stereotypes in most decision-making tasks
and shows significant cultural bias in third-party punishment and ultimatum
games. The findings indicate that, compared to humans, ChatGPT exhibits a
distinct psychological profile and cultural value orientation, and it also
shows cultural biases and stereotypes in interpersonal decision-making. Future
research endeavors should emphasize enhanced technical oversight and augmented
transparency in the database and algorithmic training procedures to foster more
efficient cross-cultural communication and mitigate social disparities. | cs.CL | The high dimensional psychological profile and cultural bias of ChatGPT |
2024-05-06T11:24:13Z | http://arxiv.org/pdf/2405.03371v2 | http://arxiv.org/abs/2405.03371v2 | Bo Wang, Jing Ma, Hongzhan Lin, Zhiwei Yang, Ruichao Yang, Yuan Tian, Yi Chang | 2024-06-20T04:33:33Z | cs.CL | Most fake news detection methods learn latent feature representations based
on neural networks, which makes them black boxes to classify a piece of news
without giving any justification. Existing explainable systems generate
veracity justifications from investigative journalism, which suffer from
debunking delayed and low efficiency. Recent studies simply assume that the
justification is equivalent to the majority opinions expressed in the wisdom of
crowds. However, the opinions typically contain some inaccurate or biased
information since the wisdom of crowds is uncensored. To detect fake news from
a sea of diverse, crowded and even competing narratives, in this paper, we
propose a novel defense-based explainable fake news detection framework.
Specifically, we first propose an evidence extraction module to split the
wisdom of crowds into two competing parties and respectively detect salient
evidences. To gain concise insights from evidences, we then design a
prompt-based module that utilizes a large language model to generate
justifications by inferring reasons towards two possible veracities. Finally,
we propose a defense-based inference module to determine veracity via modeling
the defense among these justifications. Extensive experiments conducted on two
real-world benchmarks demonstrate that our proposed method outperforms
state-of-the-art baselines in terms of fake news detection and provides
high-quality justifications. | cs.CL | Explainable Fake News Detection With Large Language Model via Defense Among Competing Wisdom |
2024-05-06T11:11:23Z | http://arxiv.org/pdf/2405.03359v1 | http://arxiv.org/abs/2405.03359v1 | Mohamed Yaseen Jabarulla, Steffen Oeltze-Jafra, Philipp Beerbaum, Theodor Uden | 2024-05-06T11:11:23Z | cs.CL, cs.AI, cs.IR | This research focuses on evaluating the non-commercial open-source large
language models (LLMs) Meditron, MedAlpaca, Mistral, and Llama-2 for their
efficacy in interpreting medical guidelines saved in PDF format. As a specific
test scenario, we applied these models to the guidelines for hypertension in
children and adolescents provided by the European Society of Cardiology (ESC).
Leveraging Streamlit, a Python library, we developed a user-friendly medical
document chatbot tool (MedDoc-Bot). This tool enables authorized users to
upload PDF files and pose questions, generating interpretive responses from
four locally stored LLMs. A pediatric expert provides a benchmark for
evaluation by formulating questions and responses extracted from the ESC
guidelines. The expert rates the model-generated responses based on their
fidelity and relevance. Additionally, we evaluated the METEOR and chrF metric
scores to assess the similarity of model responses to reference answers. Our
study found that Llama-2 and Mistral performed well in metrics evaluation.
However, Llama-2 was slower when dealing with text and tabular data. In our
human evaluation, we observed that responses created by Mistral, Meditron, and
Llama-2 exhibited reasonable fidelity and relevance. This study provides
valuable insights into the strengths and limitations of LLMs for future
developments in medical document interpretation. Open-Source Code:
https://github.com/yaseen28/MedDoc-Bot | cs.CL | MedDoc-Bot: A Chat Tool for Comparative Analysis of Large Language Models in the Context of the Pediatric Hypertension Guideline |
2024-05-06T10:52:21Z | http://arxiv.org/pdf/2405.06684v1 | http://arxiv.org/abs/2405.06684v1 | Jin Han, Zhe Zheng, Xin-Zheng Lu, Ke-Yin Chen, Jia-Rui Lin | 2024-05-06T10:52:21Z | cs.CL, cs.LG, cs.SI | Social media aids disaster response but suffers from noise, hindering
accurate impact assessment and decision making for resilient cities, which few
studies considered. To address the problem, this study proposes the first
domain-specific LLM model and an integrated method for rapid earthquake impact
assessment. First, a few categories are introduced to classify and filter
microblogs considering their relationship to the physical and social impacts of
earthquakes, and a dataset comprising 7282 earthquake-related microblogs from
twenty earthquakes in different locations is developed as well. Then, with a
systematic analysis of various influential factors, QuakeBERT, a
domain-specific large language model (LLM), is developed and fine-tuned for
accurate classification and filtering of microblogs. Meanwhile, an integrated
method integrating public opinion trend analysis, sentiment analysis, and
keyword-based physical impact quantification is introduced to assess both the
physical and social impacts of earthquakes based on social media texts.
Experiments show that data diversity and data volume dominate the performance
of QuakeBERT and increase the macro average F1 score by 27%, while the best
classification model QuakeBERT outperforms the CNN- or RNN-based models by
improving the macro average F1 score from 60.87% to 84.33%. Finally, the
proposed approach is applied to assess two earthquakes with the same magnitude
and focal depth. Results show that the proposed approach can effectively
enhance the impact assessment process by accurate detection of noisy
microblogs, which enables effective post-disaster emergency responses to create
more resilient cities. | cs.CL | QuakeBERT: Accurate Classification of Social Media Texts for Rapid Earthquake Impact Assessment |
2024-05-06T08:59:16Z | http://arxiv.org/pdf/2407.00020v1 | http://arxiv.org/abs/2407.00020v1 | Feibo Jiang, Chuanguo Tang, Li Dong, Kezhi Wang, Kun Yang, Cunhua Pan | 2024-05-06T08:59:16Z | cs.CV, cs.AI, cs.CL, cs.IT, cs.LG, math.IT | Semantic Communication (SC) has emerged as a novel communication paradigm in
recent years, successfully transcending the Shannon physical capacity limits
through innovative semantic transmission concepts. Nevertheless, extant Image
Semantic Communication (ISC) systems face several challenges in dynamic
environments, including low semantic density, catastrophic forgetting, and
uncertain Signal-to-Noise Ratio (SNR). To address these challenges, we propose
a novel Vision-Language Model-based Cross-modal Semantic Communication
(VLM-CSC) system. The VLM-CSC comprises three novel components: (1) Cross-modal
Knowledge Base (CKB) is used to extract high-density textual semantics from the
semantically sparse image at the transmitter and reconstruct the original image
based on textual semantics at the receiver. The transmission of high-density
semantics contributes to alleviating bandwidth pressure. (2) Memory-assisted
Encoder and Decoder (MED) employ a hybrid long/short-term memory mechanism,
enabling the semantic encoder and decoder to overcome catastrophic forgetting
in dynamic environments when there is a drift in the distribution of semantic
features. (3) Noise Attention Module (NAM) employs attention mechanisms to
adaptively adjust the semantic coding and the channel coding based on SNR,
ensuring the robustness of the CSC system. The experimental simulations
validate the effectiveness, adaptability, and robustness of the CSC system. | cs.CV | Visual Language Model based Cross-modal Semantic Communication Systems |
2024-05-06T08:52:11Z | http://arxiv.org/pdf/2405.03279v2 | http://arxiv.org/abs/2405.03279v2 | Qizhou Chen, Taolin Zhang, Xiaofeng He, Dongyang Li, Chengyu Wang, Longtao Huang, Hui Xue | 2024-05-08T03:45:51Z | cs.CL | Model editing aims to correct outdated or erroneous knowledge in large
language models (LLMs) without the need for costly retraining. Lifelong model
editing is the most challenging task that caters to the continuous editing
requirements of LLMs. Prior works primarily focus on single or batch editing;
nevertheless, these methods fall short in lifelong editing scenarios due to
catastrophic knowledge forgetting and the degradation of model performance.
Although retrieval-based methods alleviate these issues, they are impeded by
slow and cumbersome processes of integrating the retrieved knowledge into the
model. In this work, we introduce RECIPE, a RetriEval-augmented ContInuous
Prompt lEarning method, to boost editing efficacy and inference efficiency in
lifelong learning. RECIPE first converts knowledge statements into short and
informative continuous prompts, prefixed to the LLM's input query embedding, to
efficiently refine the response grounded on the knowledge. It further
integrates the Knowledge Sentinel (KS) that acts as an intermediary to
calculate a dynamic threshold, determining whether the retrieval repository
contains relevant knowledge. Our retriever and prompt encoder are jointly
trained to achieve editing properties, i.e., reliability, generality, and
locality. In our experiments, RECIPE is assessed extensively across multiple
LLMs and editing datasets, where it achieves superior editing performance.
RECIPE also demonstrates its capability to maintain the overall performance of
LLMs alongside showcasing fast editing and inference speed. | cs.CL | Lifelong Knowledge Editing for LLMs with Retrieval-Augmented Continuous Prompt Learning |
2024-05-06T07:12:45Z | http://arxiv.org/pdf/2405.03207v1 | http://arxiv.org/abs/2405.03207v1 | Raphaël Millière, Cameron Buckner | 2024-05-06T07:12:45Z | cs.CL | In this paper, the second of two companion pieces, we explore novel
philosophical questions raised by recent progress in large language models
(LLMs) that go beyond the classical debates covered in the first part. We focus
particularly on issues related to interpretability, examining evidence from
causal intervention methods about the nature of LLMs' internal representations
and computations. We also discuss the implications of multimodal and modular
extensions of LLMs, recent debates about whether such systems may meet minimal
criteria for consciousness, and concerns about secrecy and reproducibility in
LLM research. Finally, we discuss whether LLM-like systems may be relevant to
modeling aspects of human cognition, if their architectural characteristics and
learning scenario are adequately constrained. | cs.CL | A Philosophical Introduction to Language Models - Part II: The Way Forward |
2024-05-06T07:12:22Z | http://arxiv.org/pdf/2405.03206v1 | http://arxiv.org/abs/2405.03206v1 | Quang-Dan Tran, Van-Quan Nguyen, Quang-Huy Pham, K. B. Thang Nguyen, Trong-Hop Do | 2024-05-06T07:12:22Z | cs.CL, cs.AI | In recent years, Large Language Models (LLMs) have become integrated into our
daily lives, serving as invaluable assistants in completing tasks. Widely
embraced by users, the abuse of LLMs is inevitable, particularly in using them
to generate text content for various purposes, leading to difficulties in
distinguishing between text generated by LLMs and that written by humans. In
this study, we present a dataset named ViDetect, comprising 6.800 samples of
Vietnamese essay, with 3.400 samples authored by humans and the remainder
generated by LLMs, serving the purpose of detecting text generated by AI. We
conducted evaluations using state-of-the-art methods, including ViT5, BartPho,
PhoBERT, mDeberta V3, and mBERT. These results contribute not only to the
growing body of research on detecting text generated by AI but also demonstrate
the adaptability and effectiveness of different methods in the Vietnamese
language context. This research lays the foundation for future advancements in
AI-generated text detection and provides valuable insights for researchers in
the field of natural language processing. | cs.CL | Vietnamese AI Generated Text Detection |
2024-05-06T07:10:09Z | http://arxiv.org/pdf/2405.03205v2 | http://arxiv.org/abs/2405.03205v2 | Ruizhe Li, Yanjun Gao | 2024-05-23T07:47:02Z | cs.CL, cs.AI, cs.LG | Large Language Models (LLMs), such as the GPT-4 and LLaMA families, have
demonstrated considerable success across diverse tasks, including
multiple-choice questions (MCQs). However, these models exhibit a positional
bias, particularly an even worse anchored bias in the GPT-2 family, where they
consistently favour the first choice 'A' in MCQs during inference. This
anchored bias challenges the integrity of GPT-2's decision-making process, as
it skews performance based on the position rather than the content of the
choices in MCQs. In this study, we utilise the mechanistic interpretability
approach to identify the internal modules within GPT-2 models responsible for
this bias. We focus on the Multi-Layer Perceptron (MLP) layers and attention
heads, using the "logit lens" method to trace and modify the specific value
vectors that contribute to the bias. By updating these vectors within MLP and
recalibrating attention patterns to neutralise the preference for the first
choice 'A', we effectively mitigate the anchored bias. Our interventions not
only mitigate the bias but also improve the overall MCQ prediction accuracy for
the GPT-2 family across various datasets. This work represents the first
comprehensive mechanistic analysis of anchored bias in MCQs within the GPT-2
models, introducing targeted, minimal-intervention strategies that
significantly enhance GPT2 model robustness and accuracy in MCQs. Our code is
available at https://github.com/ruizheliUOA/Anchored_Bias_GPT2. | cs.CL | Anchored Answers: Unravelling Positional Bias in GPT-2's Multiple-Choice Questions |
2024-05-06T05:36:29Z | http://arxiv.org/pdf/2405.03170v1 | http://arxiv.org/abs/2405.03170v1 | Yueling Jenny Zeng, Li-C. Wang, Thomas Ibbetson | 2024-05-06T05:36:29Z | cs.CL | This work presents a novel approach called oracle-checker scheme for
evaluating the answer given by a generative large language model (LLM). Two
types of checkers are presented. The first type of checker follows the idea of
property testing. The second type of checker follows the idea of program
checking. Their applications are demonstrated in two separate contexts, entity
extraction and paraphrase decision, respectively. | cs.CL | Oracle-Checker Scheme for Evaluating a Generative Large Language Model |
2024-05-06T04:44:22Z | http://arxiv.org/pdf/2405.03162v1 | http://arxiv.org/abs/2405.03162v1 | Lin Yang, Shawn Xu, Andrew Sellergren, Timo Kohlberger, Yuchen Zhou, Ira Ktena, Atilla Kiraly, Faruk Ahmed, Farhad Hormozdiari, Tiam Jaroensri, Eric Wang, Ellery Wulczyn, Fayaz Jamil, Theo Guidroz, Chuck Lau, Siyuan Qiao, Yun Liu, Akshay Goel, Kendall Park, Arnav Agharwal, Nick George, Yang Wang, Ryutaro Tanno, David G. T. Barrett, Wei-Hung Weng, S. Sara Mahdavi, Khaled Saab, Tao Tu, Sreenivasa Raju Kalidindi, Mozziyar Etemadi, Jorge Cuadros, Gregory Sorensen, Yossi Matias, Katherine Chou, Greg Corrado, Joelle Barral, Shravya Shetty, David Fleet, S. M. Ali Eslami, Daniel Tse, Shruthi Prabhakara, Cory McLean, Dave Steiner, Rory Pilgrim, Christopher Kelly, Shekoofeh Azizi, Daniel Golden | 2024-05-06T04:44:22Z | cs.CV, cs.AI, cs.CL, cs.LG | Many clinical tasks require an understanding of specialized data, such as
medical images and genomics, which is not typically found in general-purpose
large multimodal models. Building upon Gemini's multimodal models, we develop
several models within the new Med-Gemini family that inherit core capabilities
of Gemini and are optimized for medical use via fine-tuning with 2D and 3D
radiology, histopathology, ophthalmology, dermatology and genomic data.
Med-Gemini-2D sets a new standard for AI-based chest X-ray (CXR) report
generation based on expert evaluation, exceeding previous best results across
two separate datasets by an absolute margin of 1% and 12%, where 57% and 96% of
AI reports on normal cases, and 43% and 65% on abnormal cases, are evaluated as
"equivalent or better" than the original radiologists' reports. We demonstrate
the first ever large multimodal model-based report generation for 3D computed
tomography (CT) volumes using Med-Gemini-3D, with 53% of AI reports considered
clinically acceptable, although additional research is needed to meet expert
radiologist reporting quality. Beyond report generation, Med-Gemini-2D
surpasses the previous best performance in CXR visual question answering (VQA)
and performs well in CXR classification and radiology VQA, exceeding SoTA or
baselines on 17 of 20 tasks. In histopathology, ophthalmology, and dermatology
image classification, Med-Gemini-2D surpasses baselines across 18 out of 20
tasks and approaches task-specific model performance. Beyond imaging,
Med-Gemini-Polygenic outperforms the standard linear polygenic risk score-based
approach for disease risk prediction and generalizes to genetically correlated
diseases for which it has never been trained. Although further development and
evaluation are necessary in the safety-critical medical domain, our results
highlight the potential of Med-Gemini across a wide range of medical tasks. | cs.CV | Advancing Multimodal Medical Capabilities of Gemini |
2024-05-06T04:42:18Z | http://arxiv.org/pdf/2405.06683v1 | http://arxiv.org/abs/2405.06683v1 | Yunxiao Shi, Xing Zi, Zijing Shi, Haimin Zhang, Qiang Wu, Min Xu | 2024-05-06T04:42:18Z | cs.CL | Retrieval-augmented generation (RAG) for language models significantly
improves language understanding systems. The basic retrieval-then-read pipeline
of response generation has evolved into a more extended process due to the
integration of various components, sometimes even forming loop structures.
Despite its advancements in improving response accuracy, challenges like poor
retrieval quality for complex questions that require the search of multifaceted
semantic information, inefficiencies in knowledge re-retrieval during long-term
serving, and lack of personalized responses persist. Motivated by transcending
these limitations, we introduce ERAGent, a cutting-edge framework that embodies
an advancement in the RAG area. Our contribution is the introduction of the
synergistically operated module: Enhanced Question Rewriter and Knowledge
Filter, for better retrieval quality. Retrieval Trigger is incorporated to
curtail extraneous external knowledge retrieval without sacrificing response
quality. ERAGent also personalizes responses by incorporating a learned user
profile. The efficiency and personalization characteristics of ERAGent are
supported by the Experiential Learner module which makes the AI assistant being
capable of expanding its knowledge and modeling user profile incrementally.
Rigorous evaluations across six datasets and three question-answering tasks
prove ERAGent's superior accuracy, efficiency, and personalization, emphasizing
its potential to advance the RAG field and its applicability in practical
systems. | cs.CL | ERAGent: Enhancing Retrieval-Augmented Language Models with Improved Accuracy, Efficiency, and Personalization |
2024-05-06T04:06:45Z | http://arxiv.org/pdf/2405.03153v1 | http://arxiv.org/abs/2405.03153v1 | Md Main Uddin Rony, Md Mahfuzul Haque, Mohammad Ali, Ahmed Shatil Alam, Naeemul Hassan | 2024-05-06T04:06:45Z | cs.CL, cs.CY, cs.LG | In the digital age, the prevalence of misleading news headlines poses a
significant challenge to information integrity, necessitating robust detection
mechanisms. This study explores the efficacy of Large Language Models (LLMs) in
identifying misleading versus non-misleading news headlines. Utilizing a
dataset of 60 articles, sourced from both reputable and questionable outlets
across health, science & tech, and business domains, we employ three LLMs-
ChatGPT-3.5, ChatGPT-4, and Gemini-for classification. Our analysis reveals
significant variance in model performance, with ChatGPT-4 demonstrating
superior accuracy, especially in cases with unanimous annotator agreement on
misleading headlines. The study emphasizes the importance of human-centered
evaluation in developing LLMs that can navigate the complexities of
misinformation detection, aligning technical proficiency with nuanced human
judgment. Our findings contribute to the discourse on AI ethics, emphasizing
the need for models that are not only technically advanced but also ethically
aligned and sensitive to the subtleties of human interpretation. | cs.CL | Exploring the Potential of the Large Language Models (LLMs) in Identifying Misleading News Headlines |
2024-05-06T03:42:34Z | http://arxiv.org/pdf/2405.03146v2 | http://arxiv.org/abs/2405.03146v2 | Sher Badshah, Hassan Sajjad | 2024-05-08T02:10:36Z | cs.LG, cs.AI, cs.CL | Scale is often attributed as one of the factors that cause an increase in the
performance of LLMs, resulting in models with billion and trillion parameters.
One of the limitations of such large models is the high computational
requirements that limit their usage, deployment, and debugging in
resource-constrained scenarios. Two commonly used alternatives to bypass these
limitations are to use the smaller versions of LLMs (e.g. Llama 7B instead of
Llama 70B) and lower the memory requirements by using quantization. While these
approaches effectively address the limitation of resources, their impact on
model performance needs thorough examination. In this study, we perform a
comprehensive evaluation to investigate the effect of model scale and
quantization on the performance. We experiment with two major families of
open-source instruct models ranging from 7 billion to 70 billion parameters.
Our extensive zero-shot experiments across various tasks including natural
language understanding, reasoning, misinformation detection, and hallucination
reveal that larger models generally outperform their smaller counterparts,
suggesting that scale remains an important factor in enhancing performance. We
found that larger models show exceptional resilience to precision reduction and
can maintain high accuracy even at 4-bit quantization for numerous tasks and
they serve as a better solution than using smaller models at high precision
under similar memory requirements. | cs.LG | Quantifying the Capabilities of LLMs across Scale and Precision |
2024-05-06T03:21:55Z | http://arxiv.org/pdf/2405.03138v1 | http://arxiv.org/abs/2405.03138v1 | Bin Wang, Geyu Lin, Zhengyuan Liu, Chengwei Wei, Nancy F. Chen | 2024-05-06T03:21:55Z | cs.CL | Large language models (LLMs) have rapidly evolved as the foundation of
various natural language processing (NLP) applications. Despite their wide use
cases, their understanding of culturally-related concepts and reasoning remains
limited. Meantime, there is a significant need to enhance these models'
cultural reasoning capabilities, especially concerning underrepresented
regions. This paper introduces a novel pipeline for extracting high-quality,
culturally-related instruction tuning datasets from vast unstructured corpora.
We utilize a self-instruction generation pipeline to identify cultural concepts
and trigger instruction. By integrating with a general-purpose instruction
tuning dataset, our model demonstrates enhanced capabilities in recognizing and
understanding regional cultural nuances, thereby enhancing its reasoning
capabilities. We conduct experiments across three regions: Singapore, the
Philippines, and the United States, achieving performance improvement of up to
6%. Our research opens new avenues for extracting cultural instruction tuning
sets directly from unstructured data, setting a precedent for future
innovations in the field. | cs.CL | CRAFT: Extracting and Tuning Cultural Instructions from the Wild |
2024-05-06T03:06:33Z | http://arxiv.org/pdf/2405.03133v1 | http://arxiv.org/abs/2405.03133v1 | Zexuan Zhong, Mengzhou Xia, Danqi Chen, Mike Lewis | 2024-05-06T03:06:33Z | cs.CL, cs.LG | Mixture-of-experts (MoE) models facilitate efficient scaling; however,
training the router network introduces the challenge of optimizing a
non-differentiable, discrete objective. Recently, a fully-differentiable MoE
architecture, SMEAR, was proposed (Muqeeth et al., 2023), which softly merges
experts in the parameter space; nevertheless, its effectiveness was only
demonstrated in downstream fine-tuning on classification tasks. In this paper,
we present Lory, the first approach that scales such architectures to
autoregressive language model pre-training. Lory introduces two key techniques:
(1) a causal segment routing strategy that achieves high efficiency for expert
merging operations while preserving the autoregressive nature of language
models; (2) a similarity-based data batching method that encourages expert
specialization by grouping similar documents in training instances. We
pre-train a series of Lory models on 150B tokens from scratch, with up to 32
experts and 30B (1.5B active) parameters. Experimental results show significant
performance gains over parameter-matched dense models on both perplexity
(+13.9%) and a variety of downstream tasks (+1.5%-11.1%). Despite segment-level
routing, Lory models achieve competitive performance compared to
state-of-the-art MoE models with token-level routing. We further demonstrate
that the trained experts in Lory capture domain-level specialization without
supervision. Our work highlights the potential of fully-differentiable MoE
architectures for language model pre-training and advocates future research in
this area. | cs.CL | Lory: Fully Differentiable Mixture-of-Experts for Autoregressive Language Model Pre-training |
2024-05-06T02:07:13Z | http://arxiv.org/pdf/2405.03111v1 | http://arxiv.org/abs/2405.03111v1 | Michael Carl | 2024-05-06T02:07:13Z | cs.CL | In this paper, we propose modelling human translation production as a
hierarchy of three embedded translation processes. The proposed architecture
replicates the temporal dynamics of keystroke production across sensorimotor,
cognitive, and phenomenal layers. Utilizing data from the CRITT TPR-DB, the
Task Segment Framework, and the HOF taxonomy, we demonstrate the temporal
breakdown of the typing flow on distinct timelines within these three layers. | cs.CL | An Active Inference Agent for Simulating Human Translation Processes in a Hierarchical Architecture: Integrating the Task Segment Framework and the HOF taxonomy |
2024-05-06T01:23:07Z | http://arxiv.org/pdf/2405.03098v1 | http://arxiv.org/abs/2405.03098v1 | Yanhong Bai, Jiabao Zhao, Jinxin Shi, Zhentao Xie, Xingjiao Wu, Liang He | 2024-05-06T01:23:07Z | cs.CL | Detecting stereotypes and biases in Large Language Models (LLMs) is crucial
for enhancing fairness and reducing adverse impacts on individuals or groups
when these models are applied. Traditional methods, which rely on embedding
spaces or are based on probability metrics, fall short in revealing the nuanced
and implicit biases present in various contexts. To address this challenge, we
propose the FairMonitor framework and adopt a static-dynamic detection method
for a comprehensive evaluation of stereotypes and biases in LLMs. The static
component consists of a direct inquiry test, an implicit association test, and
an unknown situation test, including 10,262 open-ended questions with 9
sensitive factors and 26 educational scenarios. And it is effective for
evaluating both explicit and implicit biases. Moreover, we utilize the
multi-agent system to construst the dynamic scenarios for detecting subtle
biases in more complex and realistic setting. This component detects the biases
based on the interaction behaviors of LLMs across 600 varied educational
scenarios. The experimental results show that the cooperation of static and
dynamic methods can detect more stereotypes and biased in LLMs. | cs.CL | FairMonitor: A Dual-framework for Detecting Stereotypes and Biases in Large Language Models |
2024-05-06T01:21:50Z | http://arxiv.org/pdf/2405.03097v1 | http://arxiv.org/abs/2405.03097v1 | George-Octavian Barbulescu, Peter Triantafillou | 2024-05-06T01:21:50Z | cs.LG, cs.AI, cs.CL | LLMs have been found to memorize training textual sequences and regurgitate
verbatim said sequences during text generation time. This fact is known to be
the cause of privacy and related (e.g., copyright) problems. Unlearning in LLMs
then takes the form of devising new algorithms that will properly deal with
these side-effects of memorized data, while not hurting the model's utility. We
offer a fresh perspective towards this goal, namely, that each textual sequence
to be forgotten should be treated differently when being unlearned based on its
degree of memorization within the LLM. We contribute a new metric for measuring
unlearning quality, an adversarial attack showing that SOTA algorithms lacking
this perspective fail for privacy, and two new unlearning methods based on
Gradient Ascent and Task Arithmetic, respectively. A comprehensive performance
evaluation across an extensive suite of NLP tasks then mapped the solution
space, identifying the best solutions under different scales in model
capacities and forget set sizes and quantified the gains of the new approaches. | cs.LG | To Each (Textual Sequence) Its Own: Improving Memorized-Data Unlearning in Large Language Models |
2024-05-06T00:18:43Z | http://arxiv.org/pdf/2405.03085v1 | http://arxiv.org/abs/2405.03085v1 | Kaize Shi, Xueyao Sun, Qing Li, Guandong Xu | 2024-05-06T00:18:43Z | cs.CL | Large Language Models (LLMs) have made significant strides in information
acquisition. However, their overreliance on potentially flawed parametric
knowledge leads to hallucinations and inaccuracies, particularly when handling
long-tail, domain-specific queries. Retrieval Augmented Generation (RAG)
addresses this limitation by incorporating external, non-parametric knowledge.
Nevertheless, the retrieved long-context documents often contain noisy,
irrelevant information alongside vital knowledge, negatively diluting LLMs'
attention. Inspired by the supportive role of essential concepts in
individuals' reading comprehension, we propose a novel concept-based RAG
framework with the Abstract Meaning Representation (AMR)-based concept
distillation algorithm. The proposed algorithm compresses the cluttered raw
retrieved documents into a compact set of crucial concepts distilled from the
informative nodes of AMR by referring to reliable linguistic features. The
concepts explicitly constrain LLMs to focus solely on vital information in the
inference process. We conduct extensive experiments on open-domain
question-answering datasets to empirically evaluate the proposed method's
effectiveness. The results indicate that the concept-based RAG framework
outperforms other baseline methods, particularly as the number of supporting
documents increases, while also exhibiting robustness across various backbone
LLMs. This emphasizes the distilled concepts are informative for augmenting the
RAG process by filtering out interference information. To the best of our
knowledge, this is the first work introducing AMR to enhance the RAG,
presenting a potential solution to augment inference performance with
semantic-based context compression. | cs.CL | Compressing Long Context for Enhancing RAG with AMR-based Concept Distillation |
2024-05-06T00:18:35Z | http://arxiv.org/pdf/2405.03084v1 | http://arxiv.org/abs/2405.03084v1 | Moein Shahiki Tash, Zahra Ahani, Olga Kolesnikova, Grigori Sidorov | 2024-05-06T00:18:35Z | cs.CL, cs.LG | This study delves into the relationship between emotional trends from X
platform data and the market dynamics of well-known cryptocurrencies Cardano,
Binance, Fantom, Matic, and Ripple over the period from October 2022 to March
2023. Leveraging SenticNet, we identified emotions like Fear and Anxiety, Rage
and Anger, Grief and Sadness, Delight and Pleasantness, Enthusiasm and
Eagerness, and Delight and Joy. Following data extraction, we segmented each
month into bi-weekly intervals, replicating this process for price data
obtained from Finance-Yahoo. Consequently, a comparative analysis was
conducted, establishing connections between emotional trends observed across
bi-weekly intervals and cryptocurrency prices, uncovering significant
correlations between emotional sentiments and coin valuations. | cs.CL | Analyzing Emotional Trends from X platform using SenticNet: A Comparative Analysis with Cryptocurrency Price |
2024-05-05T18:56:46Z | http://arxiv.org/pdf/2405.06682v1 | http://arxiv.org/abs/2405.06682v1 | Matthew Renze, Erhan Guven | 2024-05-05T18:56:46Z | cs.CL, cs.AI | In this study, we investigated the effects of self-reflection in large
language models (LLMs) on problem-solving performance. We instructed nine
popular LLMs to answer a series of multiple-choice questions to provide a
performance baseline. For each incorrectly answered question, we instructed
eight types of self-reflecting LLM agents to reflect on their mistakes and
provide themselves with guidance to improve problem-solving. Then, using this
guidance, each self-reflecting agent attempted to re-answer the same questions.
Our results indicate that LLM agents are able to significantly improve their
problem-solving performance through self-reflection ($p < 0.001$). In addition,
we compared the various types of self-reflection to determine their individual
contribution to performance. All code and data are available on GitHub at
https://github.com/matthewrenze/self-reflection | cs.CL | Self-Reflection in LLM Agents: Effects on Problem-Solving Performance |
2024-05-05T18:32:06Z | http://arxiv.org/pdf/2405.06681v1 | http://arxiv.org/abs/2405.06681v1 | Sven Jacobs, Steffen Jaschke | 2024-05-05T18:32:06Z | cs.CL, cs.AI, cs.CY | This paper presents the use of Retrieval Augmented Generation (RAG) to
improve the feedback generated by Large Language Models for programming tasks.
For this purpose, corresponding lecture recordings were transcribed and made
available to the Large Language Model GPT-4 as external knowledge source
together with timestamps as metainformation by using RAG. The purpose of this
is to prevent hallucinations and to enforce the use of the technical terms and
phrases from the lecture. In an exercise platform developed to solve
programming problems for an introductory programming lecture, students can
request feedback on their solutions generated by GPT-4. For this task GPT-4
receives the students' code solution, the compiler output, the result of unit
tests and the relevant passages from the lecture notes available through the
use of RAG as additional context. The feedback generated by GPT-4 should guide
students to solve problems independently and link to the lecture content, using
the time stamps of the transcript as meta-information. In this way, the
corresponding lecture videos can be viewed immediately at the corresponding
positions. For the evaluation, students worked with the tool in a workshop and
decided for each feedback whether it should be extended by RAG or not. First
results based on a questionnaire and the collected usage data show that the use
of RAG can improve feedback generation and is preferred by students in some
situations. Due to the slower speed of feedback generation, the benefits are
situation dependent. | cs.CL | Leveraging Lecture Content for Improved Feedback: Explorations with GPT-4 and Retrieval Augmented Generation |
2024-05-05T17:28:54Z | http://arxiv.org/pdf/2405.03007v1 | http://arxiv.org/abs/2405.03007v1 | Matteo Ottaviani, Stephan Stahlschmidt | 2024-05-05T17:28:54Z | cs.DL, cs.AI, cs.CL | Large bibliometric databases, such as Web of Science, Scopus, and OpenAlex,
facilitate bibliometric analyses, but are performative, affecting the
visibility of scientific outputs and the impact measurement of participating
entities. Recently, these databases have taken up the UN's Sustainable
Development Goals (SDGs) in their respective classifications, which have been
criticised for their diverging nature. This work proposes using the feature of
large language models (LLMs) to learn about the "data bias" injected by diverse
SDG classifications into bibliometric data by exploring five SDGs. We build a
LLM that is fine-tuned in parallel by the diverse SDG classifications inscribed
into the databases' SDG classifications. Our results show high sensitivity in
model architecture, classified publications, fine-tuning process, and natural
language generation. The wide arbitrariness at different levels raises concerns
about using LLM in research practice. | cs.DL | On the performativity of SDG classifications in large bibliometric databases |
2024-05-05T17:19:35Z | http://arxiv.org/pdf/2405.03004v1 | http://arxiv.org/abs/2405.03004v1 | Yuxi Xia, Anastasiia Sedova, Pedro Henrique Luz de Araujo, Vasiliki Kougia, Lisa Nußbaumer, Benjamin Roth | 2024-05-05T17:19:35Z | cs.CL, cs.LG | Training data memorization in language models impacts model capability
(generalization) and safety (privacy risk). This paper focuses on analyzing
prompts' impact on detecting the memorization of 6 masked language model-based
named entity recognition models. Specifically, we employ a diverse set of 400
automatically generated prompts, and a pairwise dataset where each pair
consists of one person's name from the training set and another name out of the
set. A prompt completed with a person's name serves as input for getting the
model's confidence in predicting this name. Finally, the prompt performance of
detecting model memorization is quantified by the percentage of name pairs for
which the model has higher confidence for the name from the training set. We
show that the performance of different prompts varies by as much as 16
percentage points on the same model, and prompt engineering further increases
the gap. Moreover, our experiments demonstrate that prompt performance is
model-dependent but does generalize across different name sets. A comprehensive
analysis indicates how prompt performance is influenced by prompt properties,
contained tokens, and the model's self-attention weights on the prompt. | cs.CL | Exploring prompts to elicit memorization in masked language model-based named entity recognition |
2024-05-05T17:15:24Z | http://arxiv.org/pdf/2405.03003v1 | http://arxiv.org/abs/2405.03003v1 | Ziqi Gao, Qichao Wang, Aochuan Chen, Zijing Liu, Bingzhe Wu, Liang Chen, Jia Li | 2024-05-05T17:15:24Z | cs.LG, cs.AI, cs.CL | Low-rank adaptation~(LoRA) has recently gained much interest in fine-tuning
foundation models. It effectively reduces the number of trainable parameters by
incorporating low-rank matrices $A$ and $B$ to represent the weight change,
i.e., $\Delta W=BA$. Despite LoRA's progress, it faces storage challenges when
handling extensive customization adaptations or larger base models. In this
work, we aim to further compress trainable parameters by enjoying the powerful
expressiveness of the Fourier transform. Specifically, we introduce FourierFT,
which treats $\Delta W$ as a matrix in the spatial domain and learns only a
small fraction of its spectral coefficients. With the trained spectral
coefficients, we implement the inverse discrete Fourier transform to recover
$\Delta W$. Empirically, our FourierFT method shows comparable or better
performance with fewer parameters than LoRA on various tasks, including natural
language understanding, natural language generation, instruction tuning, and
image classification. For example, when performing instruction tuning on the
LLaMA2-7B model, FourierFT surpasses LoRA with only 0.064M trainable
parameters, compared to LoRA's 33.5M. Our code is released at
\url{https://github.com/Chaos96/fourierft}. | cs.LG | Parameter-Efficient Fine-Tuning with Discrete Fourier Transform |
2024-05-05T17:06:31Z | http://arxiv.org/pdf/2405.03000v1 | http://arxiv.org/abs/2405.03000v1 | Wenqi Shi, Ran Xu, Yuchen Zhuang, Yue Yu, Hang Wu, Carl Yang, May D. Wang | 2024-05-05T17:06:31Z | cs.CL, cs.AI | Despite their improved capabilities in generation and reasoning, adapting
large language models (LLMs) to the biomedical domain remains challenging due
to their immense size and corporate privacy. In this work, we propose
MedAdapter, a unified post-hoc adapter for test-time adaptation of LLMs towards
biomedical applications. Instead of fine-tuning the entire LLM, MedAdapter
effectively adapts the original model by fine-tuning only a small BERT-sized
adapter to rank candidate solutions generated by LLMs. Experiments demonstrate
that MedAdapter effectively adapts both white-box and black-box LLMs in
biomedical reasoning, achieving average performance improvements of 25.48% and
11.31%, respectively, without requiring extensive computational resources or
sharing data with third parties. MedAdapter also yields superior performance
when combined with train-time adaptation, highlighting a flexible and
complementary solution to existing adaptation methods. Faced with the
challenges of balancing model performance, computational resources, and data
privacy, MedAdapter provides an efficient, privacy-preserving, cost-effective,
and transparent solution for adapting LLMs to the biomedical domain. | cs.CL | MedAdapter: Efficient Test-Time Adaptation of Large Language Models towards Medical Reasoning |
2024-05-05T16:35:30Z | http://arxiv.org/pdf/2405.06680v1 | http://arxiv.org/abs/2405.06680v1 | Jun Zhao, Jingqi Tong, Yurong Mou, Ming Zhang, Qi Zhang, Xuanjing Huang | 2024-05-05T16:35:30Z | cs.CL, cs.AI | Human cognition exhibits systematic compositionality, the algebraic ability
to generate infinite novel combinations from finite learned components, which
is the key to understanding and reasoning about complex logic. In this work, we
investigate the compositionality of large language models (LLMs) in
mathematical reasoning. Specifically, we construct a new dataset
\textsc{MathTrap}\footnotemark[3] by introducing carefully designed logical
traps into the problem descriptions of MATH and GSM8k. Since problems with
logical flaws are quite rare in the real world, these represent ``unseen''
cases to LLMs. Solving these requires the models to systematically compose (1)
the mathematical knowledge involved in the original problems with (2) knowledge
related to the introduced traps. Our experiments show that while LLMs possess
both components of requisite knowledge, they do not \textbf{spontaneously}
combine them to handle these novel cases. We explore several methods to
mitigate this deficiency, such as natural language prompts, few-shot
demonstrations, and fine-tuning. We find that LLMs' performance can be
\textbf{passively} improved through the above external intervention. Overall,
systematic compositionality remains an open challenge for large language
models. | cs.CL | Exploring the Compositional Deficiency of Large Language Models in Mathematical Reasoning |
2024-05-05T16:11:06Z | http://arxiv.org/pdf/2405.02985v1 | http://arxiv.org/abs/2405.02985v1 | Owen Henkel, Adam Boxer, Libby Hills, Bill Roberts | 2024-05-05T16:11:06Z | cs.CL, cs.AI | This paper presents reports on a series of experiments with a novel dataset
evaluating how well Large Language Models (LLMs) can mark (i.e. grade) open
text responses to short answer questions, Specifically, we explore how well
different combinations of GPT version and prompt engineering strategies
performed at marking real student answers to short answer across different
domain areas (Science and History) and grade-levels (spanning ages 5-16) using
a new, never-used-before dataset from Carousel, a quizzing platform. We found
that GPT-4, with basic few-shot prompting performed well (Kappa, 0.70) and,
importantly, very close to human-level performance (0.75). This research builds
on prior findings that GPT-4 could reliably score short answer reading
comprehension questions at a performance-level very close to that of expert
human raters. The proximity to human-level performance, across a variety of
subjects and grade levels suggests that LLMs could be a valuable tool for
supporting low-stakes formative assessment tasks in K-12 education and has
important implications for real-world education delivery. | cs.CL | Can Large Language Models Make the Grade? An Empirical Study Evaluating LLMs Ability to Mark Short Answer Questions in K-12 Education |
2024-05-05T16:07:23Z | http://arxiv.org/pdf/2405.02984v1 | http://arxiv.org/abs/2405.02984v1 | Şükrü Öztürk, Hacer Yalim Keles | 2024-05-05T16:07:23Z | cs.CL, cs.CV, cs.LG | This study introduces the continuous Educational Turkish Sign Language
(E-TSL) dataset, collected from online Turkish language lessons for 5th, 6th,
and 8th grades. The dataset comprises 1,410 videos totaling nearly 24 hours and
includes performances from 11 signers. Turkish, an agglutinative language,
poses unique challenges for sign language translation, particularly with a
vocabulary where 64% are singleton words and 85% are rare words, appearing less
than five times. We developed two baseline models to address these challenges:
the Pose to Text Transformer (P2T-T) and the Graph Neural Network based
Transformer (GNN-T) models. The GNN-T model achieved 19.13% BLEU-1 score and
3.28% BLEU-4 score, presenting a significant challenge compared to existing
benchmarks. The P2T-T model, while demonstrating slightly lower performance in
BLEU scores, achieved a higher ROUGE-L score of 22.09%. Additionally, we
benchmarked our model using the well-known PHOENIX-Weather 2014T dataset to
validate our approach. | cs.CL | E-TSL: A Continuous Educational Turkish Sign Language Dataset with Baseline Methods |
2024-05-05T13:57:05Z | http://arxiv.org/pdf/2405.02937v2 | http://arxiv.org/abs/2405.02937v2 | Fatema Tuj Johora Faria, Mukaffi Bin Moin, Asif Iftekher Fahim, Pronay Debnath, Faisal Muhammad Shah | 2024-05-07T17:34:13Z | cs.CL | Natural Language Inference (NLI) is a cornerstone of Natural Language
Processing (NLP), providing insights into the entailment relationships between
text pairings. It is a critical component of Natural Language Understanding
(NLU), demonstrating the ability to extract information from spoken or written
interactions. NLI is mainly concerned with determining the entailment
relationship between two statements, known as the premise and hypothesis. When
the premise logically implies the hypothesis, the pair is labeled "entailment".
If the hypothesis contradicts the premise, the pair receives the
"contradiction" label. When there is insufficient evidence to establish a
connection, the pair is described as "neutral". Despite the success of Large
Language Models (LLMs) in various tasks, their effectiveness in NLI remains
constrained by issues like low-resource domain accuracy, model overconfidence,
and difficulty in capturing human judgment disagreements. This study addresses
the underexplored area of evaluating LLMs in low-resourced languages such as
Bengali. Through a comprehensive evaluation, we assess the performance of
prominent LLMs and state-of-the-art (SOTA) models in Bengali NLP tasks,
focusing on natural language inference. Utilizing the XNLI dataset, we conduct
zero-shot and few-shot evaluations, comparing LLMs like GPT-3.5 Turbo and
Gemini 1.5 Pro with models such as BanglaBERT, Bangla BERT Base, DistilBERT,
mBERT, and sahajBERT. Our findings reveal that while LLMs can achieve
comparable or superior performance to fine-tuned SOTA models in few-shot
scenarios, further research is necessary to enhance our understanding of LLMs
in languages with modest resources like Bengali. This study underscores the
importance of continued efforts in exploring LLM capabilities across diverse
linguistic contexts. | cs.CL | Unraveling the Dominance of Large Language Models Over Transformer Models for Bangla Natural Language Inference: A Comprehensive Study |
2024-05-05T13:54:02Z | http://arxiv.org/pdf/2405.02935v1 | http://arxiv.org/abs/2405.02935v1 | Zhixiang Su, Yinan Zhang, Jiazheng Jing, Jie Xiao, Zhiqi Shen | 2024-05-05T13:54:02Z | cs.CL | Disease prediction holds considerable significance in modern healthcare,
because of its crucial role in facilitating early intervention and implementing
effective prevention measures. However, most recent disease prediction
approaches heavily rely on laboratory test outcomes (e.g., blood tests and
medical imaging from X-rays). Gaining access to such data for precise disease
prediction is often a complex task from the standpoint of a patient and is
always only available post-patient consultation. To make disease prediction
available from patient-side, we propose Personalized Medical Disease Prediction
(PoMP), which predicts diseases using patient health narratives including
textual descriptions and demographic information. By applying PoMP, patients
can gain a clearer comprehension of their conditions, empowering them to
directly seek appropriate medical specialists and thereby reducing the time
spent navigating healthcare communication to locate suitable doctors. We
conducted extensive experiments using real-world data from Haodf to showcase
the effectiveness of PoMP. | cs.CL | Enabling Patient-side Disease Prediction via the Integration of Patient Narratives |
2024-05-05T13:42:25Z | http://arxiv.org/pdf/2405.02933v1 | http://arxiv.org/abs/2405.02933v1 | Chengpeng Fu, Xiaocheng Feng, Yichong Huang, Wenshuai Huo, Baohang Li, Hui Wang, Bin Qin, Ting Liu | 2024-05-05T13:42:25Z | cs.CL | Leveraging large language models for machine translation has demonstrated
promising results. However, it does require the large language models to
possess the capability of handling both the source and target languages in
machine translation. When it is challenging to find large models that support
the desired languages, resorting to continuous learning methods becomes a
costly endeavor. To mitigate these expenses, we propose an innovative approach
called RD (Relay Decoding), which entails concatenating two distinct large
models that individually support the source and target languages. By
incorporating a simple mapping layer to facilitate the connection between these
two models and utilizing a limited amount of parallel data for training, we
successfully achieve superior results in the machine translation task.
Experimental results conducted on the Multi30k and WikiMatrix datasets validate
the effectiveness of our proposed method. | cs.CL | Relay Decoding: Concatenating Large Language Models for Machine Translation |
2024-05-05T13:09:55Z | http://arxiv.org/pdf/2405.02925v1 | http://arxiv.org/abs/2405.02925v1 | Guanhua Chen, Yutong Yao, Derek F. Wong, Lidia S. Chao | 2024-05-05T13:09:55Z | cs.CL | Multi-intent natural language understanding (NLU) presents a formidable
challenge due to the model confusion arising from multiple intents within a
single utterance. While previous works train the model contrastively to
increase the margin between different multi-intent labels, they are less suited
to the nuances of multi-intent NLU. They ignore the rich information between
the shared intents, which is beneficial to constructing a better embedding
space, especially in low-data scenarios. We introduce a two-stage
Prediction-Aware Contrastive Learning (PACL) framework for multi-intent NLU to
harness this valuable knowledge. Our approach capitalizes on shared intent
information by integrating word-level pre-training and prediction-aware
contrastive fine-tuning. We construct a pre-training dataset using a word-level
data augmentation strategy. Subsequently, our framework dynamically assigns
roles to instances during contrastive fine-tuning while introducing a
prediction-aware contrastive loss to maximize the impact of contrastive
learning. We present experimental results and empirical analysis conducted on
three widely used datasets, demonstrating that our method surpasses the
performance of three prominent baselines on both low-data and full-data
scenarios. | cs.CL | A Two-Stage Prediction-Aware Contrastive Learning Framework for Multi-Intent NLU |
2024-05-05T12:51:38Z | http://arxiv.org/pdf/2405.02917v1 | http://arxiv.org/abs/2405.02917v1 | Tobias Groot, Matias Valdenegro-Toro | 2024-05-05T12:51:38Z | cs.CV, cs.CL, cs.LG | Language and Vision-Language Models (LLMs/VLMs) have revolutionized the field
of AI by their ability to generate human-like text and understand images, but
ensuring their reliability is crucial. This paper aims to evaluate the ability
of LLMs (GPT4, GPT-3.5, LLaMA2, and PaLM 2) and VLMs (GPT4V and Gemini Pro
Vision) to estimate their verbalized uncertainty via prompting. We propose the
new Japanese Uncertain Scenes (JUS) dataset, aimed at testing VLM capabilities
via difficult queries and object counting, and the Net Calibration Error (NCE)
to measure direction of miscalibration. Results show that both LLMs and VLMs
have a high calibration error and are overconfident most of the time,
indicating a poor capability for uncertainty estimation. Additionally we
develop prompts for regression tasks, and we show that VLMs have poor
calibration when producing mean/standard deviation and 95% confidence
intervals. | cs.CV | Overconfidence is Key: Verbalized Uncertainty Evaluation in Large Language and Vision-Language Models |
2024-05-05T10:52:09Z | http://arxiv.org/pdf/2405.02887v1 | http://arxiv.org/abs/2405.02887v1 | Aekansh Kathunia, Mohammad Kaif, Nalin Arora, N Narotam | 2024-05-05T10:52:09Z | cs.CL, cs.AI | People communicate in more than 7,000 languages around the world, with around
780 languages spoken in India alone. Despite this linguistic diversity,
research on Sentiment Analysis has predominantly focused on English text data,
resulting in a disproportionate availability of sentiment resources for
English. This paper examines the performance of transformer models in Sentiment
Analysis tasks across multilingual datasets and text that has undergone machine
translation. By comparing the effectiveness of these models in different
linguistic contexts, we gain insights into their performance variations and
potential implications for sentiment analysis across diverse languages. We also
discuss the shortcomings and potential for future work towards the end. | cs.CL | Sentiment Analysis Across Languages: Evaluation Before and After Machine Translation to English |
2024-05-05T09:20:38Z | http://arxiv.org/pdf/2405.02861v1 | http://arxiv.org/abs/2405.02861v1 | Yang Liu, Melissa Xiaohui Qin, Hongming Li, Chao Huang | 2024-05-05T09:20:38Z | cs.CL, cs.AI, cs.LG, 68T50, I.2.7 | We introduce LexBench, a comprehensive evaluation suite enabled to test
language models (LMs) on ten semantic phrase processing tasks. Unlike prior
studies, it is the first work to propose a framework from the comparative
perspective to model the general semantic phrase (i.e., lexical collocation)
and three fine-grained semantic phrases, including idiomatic expression, noun
compound, and verbal construction. Thanks to \ourbenchmark, we assess the
performance of 15 LMs across model architectures and parameter scales in
classification, extraction, and interpretation tasks. Through the experiments,
we first validate the scaling law and find that, as expected, large models
excel better than the smaller ones in most tasks. Second, we investigate
further through the scaling semantic relation categorization and find that
few-shot LMs still lag behind vanilla fine-tuned models in the task. Third,
through human evaluation, we find that the performance of strong models is
comparable to the human level regarding semantic phrase processing. Our
benchmarking findings can serve future research aiming to improve the generic
capability of LMs on semantic phrase comprehension. Our source code and data
are available at https://github.com/jacklanda/LexBench | cs.CL | Revisiting a Pain in the Neck: Semantic Phrase Processing Benchmark for Language Models |
2024-05-05T09:02:54Z | http://arxiv.org/pdf/2405.02858v1 | http://arxiv.org/abs/2405.02858v1 | Jinyu Cai, Jialong Li, Mingyue Zhang, Munan Li, Chen-Shu Wang, Kenji Tei | 2024-05-05T09:02:54Z | cs.SI, cs.CL | Social media platforms such as Twitter, Reddit, and Sina Weibo play a crucial
role in global communication but often encounter strict regulations in
geopolitically sensitive regions. This situation has prompted users to
ingeniously modify their way of communicating, frequently resorting to coded
language in these regulated social media environments. This shift in
communication is not merely a strategy to counteract regulation, but a vivid
manifestation of language evolution, demonstrating how language naturally
evolves under societal and technological pressures. Studying the evolution of
language in regulated social media contexts is of significant importance for
ensuring freedom of speech, optimizing content moderation, and advancing
linguistic research. This paper proposes a multi-agent simulation framework
using Large Language Models (LLMs) to explore the evolution of user language in
regulated social media environments. The framework employs LLM-driven agents:
supervisory agent who enforce dialogue supervision and participant agents who
evolve their language strategies while engaging in conversation, simulating the
evolution of communication styles under strict regulations aimed at evading
social media regulation. The study evaluates the framework's effectiveness
through a range of scenarios from abstract scenarios to real-world situations.
Key findings indicate that LLMs are capable of simulating nuanced language
dynamics and interactions in constrained settings, showing improvement in both
evading supervision and information accuracy as evolution progresses.
Furthermore, it was found that LLM agents adopt different strategies for
different scenarios. | cs.SI | Language Evolution for Evading Social Media Regulation via LLM-based Multi-agent Simulation |
2024-05-05T05:43:20Z | http://arxiv.org/pdf/2405.02817v1 | http://arxiv.org/abs/2405.02817v1 | Huanjun Kong | 2024-05-05T05:43:20Z | cs.CL | How to eliminate pronominal reference in group chats? In this work, we have
preprocessed 58k authentic chat data and manually annotated 2.3k questions. The
reliability of this annotation was confirmed by the scaling law. After this, we
conducted fine-tuning on Qwen models, ranging from 0.5B to 32B parameters. The
optimal version improved 29.07 in F1 score. This confirms the viability of
fine-tuning Large Language Model (LLM) for downstream Natural Language
Processing (NLP) tasks. Our contributions are: 1) Created Supervised
Fine-Tuning (SFT) training data in alpaca format, along with a set of Low-Rank
Adaptation (LoRA) weights, and 2) Developed a method for acquiring high-quality
data leveraging scaling law principle. The script, raw data with alpaca format
and experiments track are open-sourced on Github
https://github.com/InternLM/HuixiangDou/tree/main/web/tools, HuggingFace
https://huggingface.co/tpoisonooo and WandB
https://wandb.ai/tpoisonooo/huixiangdou-cr/table?nw=nwusertpoisonooo . The
privacy of the data involved has been authorized by users. | cs.CL | HuixiangDou-CR: Coreference Resolution in Group Chats |
2024-05-05T05:42:33Z | http://arxiv.org/pdf/2405.02816v1 | http://arxiv.org/abs/2405.02816v1 | Hamed Zamani, Michael Bendersky | 2024-05-05T05:42:33Z | cs.CL, cs.IR, cs.LG | This paper introduces Stochastic RAG--a novel approach for end-to-end
optimization of retrieval-augmented generation (RAG) models that relaxes the
simplifying assumptions of marginalization and document independence, made in
most prior work. Stochastic RAG casts the retrieval process in RAG as a
stochastic sampling without replacement process. Through this formulation, we
employ straight-through Gumbel-top-k that provides a differentiable
approximation for sampling without replacement and enables effective end-to-end
optimization for RAG. We conduct extensive experiments on seven diverse
datasets on a wide range of tasks, from open-domain question answering to fact
verification to slot-filling for relation extraction and to dialogue systems.
By applying this optimization method to a recent and effective RAG model, we
advance state-of-the-art results on six out of seven datasets. | cs.CL | Stochastic RAG: End-to-End Retrieval-Augmented Generation through Expected Utility Maximization |
2024-05-05T05:06:07Z | http://arxiv.org/pdf/2405.02814v2 | http://arxiv.org/abs/2405.02814v2 | Xu Wang, Cheng Li, Yi Chang, Jindong Wang, Yuan Wu | 2024-05-12T16:11:48Z | cs.CL | Large Language Models (LLMs) have become integral to a wide spectrum of
applications, ranging from traditional computing tasks to advanced artificial
intelligence (AI) applications. This widespread adoption has spurred extensive
research into LLMs across various disciplines, including the social sciences.
Notably, studies have revealed that LLMs possess emotional intelligence, which
can be further developed through positive emotional stimuli. This discovery
raises an intriguing question: can negative emotions similarly influence LLMs,
potentially enhancing their performance? In response to this question, we
introduce NegativePrompt, a novel approach underpinned by psychological
principles, involving ten specifically designed negative emotional stimuli. We
embark on rigorous experimental evaluations of five LLMs including
Flan-T5-Large, Vicuna, Llama 2, ChatGPT, and GPT-4, across a set of 45 tasks.
The results are revealing: NegativePrompt markedly enhances the performance of
LLMs, evidenced by relative improvements of 12.89% in Instruction Induction
tasks and 46.25% in BIG-Bench tasks. Moreover, we conduct attention
visualization experiments to decipher the underlying mechanisms of
NegativePrompt's influence. Our research contributes significantly to the
understanding of LLMs and emotion interaction, demonstrating the practical
efficacy of NegativePrompt as an emotion-driven method and offering novel
insights for the enhancement of LLMs in real-world applications. The code is
available at https://github.com/wangxu0820/NegativePrompt. | cs.CL | NegativePrompt: Leveraging Psychology for Large Language Models Enhancement via Negative Emotional Stimuli |
2024-05-05T02:15:11Z | http://arxiv.org/pdf/2405.02793v1 | http://arxiv.org/abs/2405.02793v1 | Roopal Garg, Andrea Burns, Burcu Karagol Ayan, Yonatan Bitton, Ceslee Montgomery, Yasumasa Onoe, Andrew Bunner, Ranjay Krishna, Jason Baldridge, Radu Soricut | 2024-05-05T02:15:11Z | cs.CV, cs.CL | Despite the longstanding adage "an image is worth a thousand words," creating
accurate and hyper-detailed image descriptions for training Vision-Language
models remains challenging. Current datasets typically have web-scraped
descriptions that are short, low-granularity, and often contain details
unrelated to the visual content. As a result, models trained on such data
generate descriptions replete with missing information, visual inconsistencies,
and hallucinations. To address these issues, we introduce ImageInWords (IIW), a
carefully designed human-in-the-loop annotation framework for curating
hyper-detailed image descriptions and a new dataset resulting from this
process. We validate the framework through evaluations focused on the quality
of the dataset and its utility for fine-tuning with considerations for
readability, comprehensiveness, specificity, hallucinations, and
human-likeness. Our dataset significantly improves across these dimensions
compared to recently released datasets (+66%) and GPT-4V outputs (+48%).
Furthermore, models fine-tuned with IIW data excel by +31% against prior work
along the same human evaluation dimensions. Given our fine-tuned models, we
also evaluate text-to-image generation and vision-language reasoning. Our
model's descriptions can generate images closest to the original, as judged by
both automated and human metrics. We also find our model produces more
compositionally rich descriptions, outperforming the best baseline by up to 6%
on ARO, SVO-Probes, and Winoground datasets. | cs.CV | ImageInWords: Unlocking Hyper-Detailed Image Descriptions |
2024-05-05T02:06:37Z | http://arxiv.org/pdf/2405.06677v1 | http://arxiv.org/abs/2405.06677v1 | Xiaohan Lin, Qingxing Cao, Yinya Huang, Zhicheng Yang, Zhengying Liu, Zhenguo Li, Xiaodan Liang | 2024-05-05T02:06:37Z | cs.CL, cs.AI | Humans can develop new theorems to explore broader and more complex
mathematical results. While current generative language models (LMs) have
achieved significant improvement in automatically proving theorems, their
ability to generate new or reusable theorems is still under-explored. Without
the new theorems, current LMs struggle to prove harder theorems that are
distant from the given hypotheses with the exponentially growing search space.
Therefore, this paper proposes an Automated Theorem Generation (ATG) benchmark
that evaluates whether an agent can automatically generate valuable (and
possibly brand new) theorems that are applicable for downstream theorem proving
as reusable knowledge. Specifically, we construct the ATG benchmark by
splitting the Metamath library into three sets: axioms, library, and problem
based on their proving depth. We conduct extensive experiments to investigate
whether current LMs can generate theorems in the library and benefit the
problem theorems proving. The results demonstrate that high-quality ATG data
facilitates models' performances on downstream ATP. However, there is still
room for current LMs to develop better ATG and generate more advanced and
human-like theorems. We hope the new ATG challenge can shed some light on
advanced complex theorem proving. | cs.CL | ATG: Benchmarking Automated Theorem Generation for Generative Language Models |
2024-05-05T00:08:00Z | http://arxiv.org/pdf/2405.02774v1 | http://arxiv.org/abs/2405.02774v1 | Feiyang Kang, Hoang Anh Just, Yifan Sun, Himanshu Jahagirdar, Yuanzhi Zhang, Rongxing Du, Anit Kumar Sahu, Ruoxi Jia | 2024-05-05T00:08:00Z | cs.LG, cs.AI, cs.CL | This work focuses on leveraging and selecting from vast, unlabeled, open data
to pre-fine-tune a pre-trained language model. The goal is to minimize the need
for costly domain-specific data for subsequent fine-tuning while achieving
desired performance levels. While many data selection algorithms have been
designed for small-scale applications, rendering them unsuitable for our
context, some emerging methods do cater to language data scales. However, they
often prioritize data that aligns with the target distribution. While this
strategy may be effective when training a model from scratch, it can yield
limited results when the model has already been pre-trained on a different
distribution. Differing from prior work, our key idea is to select data that
nudges the pre-training distribution closer to the target distribution. We show
the optimality of this approach for fine-tuning tasks under certain conditions.
We demonstrate the efficacy of our methodology across a diverse array of tasks
(NLU, NLG, zero-shot) with models up to 2.7B, showing that it consistently
surpasses other selection methods. Moreover, our proposed method is
significantly faster than existing techniques, scaling to millions of samples
within a single GPU hour. Our code is open-sourced (Code repository:
https://anonymous.4open.science/r/DV4LLM-D761/ ). While fine-tuning offers
significant potential for enhancing performance across diverse tasks, its
associated costs often limit its widespread adoption; with this work, we hope
to lay the groundwork for cost-effective fine-tuning, making its benefits more
accessible. | cs.LG | Get more for less: Principled Data Selection for Warming Up Fine-Tuning in LLMs |
2024-05-04T22:02:24Z | http://arxiv.org/pdf/2405.02765v2 | http://arxiv.org/abs/2405.02765v2 | Paul Youssef, Zhixue Zhao, Jörg Schlötterer, Christin Seifert | 2024-07-01T19:20:58Z | cs.CL, cs.AI | Knowledge editing methods (KEs) can update language models' obsolete or
inaccurate knowledge learned from pre-training. However, KEs can be used for
malicious applications, e.g., inserting misinformation and toxic content.
Knowing whether a generated output is based on edited knowledge or first-hand
knowledge from pre-training can increase users' trust in generative models and
provide more transparency. Driven by this, we propose a novel task: detecting
edited knowledge in language models. Given an edited model and a fact retrieved
by a prompt from an edited model, the objective is to classify the knowledge as
either unedited (based on the pre-training), or edited (based on subsequent
editing). We instantiate the task with four KEs, two LLMs, and two datasets.
Additionally, we propose using the hidden state representations and the
probability distributions as features for the detection. Our results reveal
that, using these features as inputs to a simple AdaBoost classifiers
establishes a strong baseline. This classifier requires only a limited amount
of data and maintains its performance even in cross-domain settings. Last, we
find it more challenging to distinguish edited knowledge from unedited but
related knowledge, highlighting the need for further research. Our work lays
the groundwork for addressing malicious model editing, which is a critical
challenge associated with the strong generative capabilities of LLMs. | cs.CL | Detecting Edited Knowledge in Language Models |
2024-05-04T22:00:28Z | http://arxiv.org/pdf/2405.02764v1 | http://arxiv.org/abs/2405.02764v1 | Zeyu Yang, Zhao Meng, Xiaochen Zheng, Roger Wattenhofer | 2024-05-04T22:00:28Z | cs.CL, cs.LG | Large Language Models (LLMs) have revolutionized natural language processing,
but their robustness against adversarial attacks remains a critical concern. We
presents a novel white-box style attack approach that exposes vulnerabilities
in leading open-source LLMs, including Llama, OPT, and T5. We assess the impact
of model size, structure, and fine-tuning strategies on their resistance to
adversarial perturbations. Our comprehensive evaluation across five diverse
text classification tasks establishes a new benchmark for LLM robustness. The
findings of this study have far-reaching implications for the reliable
deployment of LLMs in real-world applications and contribute to the advancement
of trustworthy AI systems. | cs.CL | Assessing Adversarial Robustness of Large Language Models: An Empirical Study |
2024-05-04T21:29:37Z | http://arxiv.org/pdf/2405.06676v1 | http://arxiv.org/abs/2405.06676v1 | Bing-Yue Wu, Utsav Sharma, Sai Rahul Dhanvi Kankipati, Ajay Yadav, Bintu Kappil George, Sai Ritish Guntupalli, Austin Rovinski, Vidya A. Chhabria | 2024-05-04T21:29:37Z | cs.CL, cs.AI, cs.AR | Large language models (LLMs) serve as powerful tools for design, providing
capabilities for both task automation and design assistance. Recent
advancements have shown tremendous potential for facilitating LLM integration
into the chip design process; however, many of these works rely on data that
are not publicly available and/or not permissively licensed for use in LLM
training and distribution. In this paper, we present a solution aimed at
bridging this gap by introducing an open-source dataset tailored for OpenROAD,
a widely adopted open-source EDA toolchain. The dataset features over 1000 data
points and is structured in two formats: (i) a pairwise set comprised of
question prompts with prose answers, and (ii) a pairwise set comprised of code
prompts and their corresponding OpenROAD scripts. By providing this dataset, we
aim to facilitate LLM-focused research within the EDA domain. The dataset is
available at https://github.com/OpenROAD-Assistant/EDA-Corpus. | cs.CL | EDA Corpus: A Large Language Model Dataset for Enhanced Interaction with OpenROAD |
2024-05-04T20:38:41Z | http://arxiv.org/pdf/2405.02750v1 | http://arxiv.org/abs/2405.02750v1 | Zheng Zhao, Emilio Monti, Jens Lehmann, Haytham Assem | 2024-05-04T20:38:41Z | cs.CL, cs.AI | Large language models (LLMs) tend to inadequately integrate input context
during text generation, relying excessively on encoded prior knowledge in model
parameters, potentially resulting in generated text with factual
inconsistencies or contextually unfaithful content. LLMs utilize two primary
knowledge sources: 1) prior (parametric) knowledge from pretraining, and 2)
contextual (non-parametric) knowledge from input prompts. The study addresses
the open question of how LLMs effectively balance these knowledge sources
during the generation process, specifically in the context of open-domain
question answering. To address this issue, we introduce a novel approach
integrating contrastive decoding with adversarial irrelevant passages as
negative samples to enhance robust context grounding during generation.
Notably, our method operates at inference time without requiring further
training. We conduct comprehensive experiments to demonstrate its applicability
and effectiveness, providing empirical evidence showcasing its superiority over
existing methodologies. Our code is publicly available at:
https://github.com/amazon-science/ContextualUnderstanding-ContrastiveDecoding. | cs.CL | Enhancing Contextual Understanding in Large Language Models through Contrastive Decoding |
2024-05-04T19:53:03Z | http://arxiv.org/pdf/2405.02743v1 | http://arxiv.org/abs/2405.02743v1 | Yuval Reif, Roy Schwartz | 2024-05-04T19:53:03Z | cs.CL | Large language models (LLMs) have shown remarkable adaptability to diverse
tasks, by leveraging context prompts containing instructions, or minimal
input-output examples. However, recent work revealed they also exhibit label
bias -- an undesirable preference toward predicting certain answers over
others. Still, detecting and measuring this bias reliably and at scale has
remained relatively unexplored. In this study, we evaluate different approaches
to quantifying label bias in a model's predictions, conducting a comprehensive
investigation across 279 classification tasks and ten LLMs. Our investigation
reveals substantial label bias in models both before and after debiasing
attempts, as well as highlights the importance of outcomes-based evaluation
metrics, which were not previously used in this regard. We further propose a
novel label bias calibration method tailored for few-shot prompting, which
outperforms recent calibration approaches for both improving performance and
mitigating label bias. Our results emphasize that label bias in the predictions
of LLMs remains a barrier to their reliability. | cs.CL | Beyond Performance: Quantifying and Mitigating Label Bias in LLMs |
2024-05-04T19:04:51Z | http://arxiv.org/pdf/2405.02738v1 | http://arxiv.org/abs/2405.02738v1 | Sakher Khalil Alqaaidi, Krzysztof Kochut | 2024-05-04T19:04:51Z | cs.CL, cs.AI, I.2.4 | Knowledge Graphs have been widely used to represent facts in a structured
format. Due to their large scale applications, knowledge graphs suffer from
being incomplete. The relation prediction task obtains knowledge graph
completion by assigning one or more possible relations to each pair of nodes.
In this work, we make use of the knowledge graph node names to fine-tune a
large language model for the relation prediction task. By utilizing the node
names only we enable our model to operate sufficiently in the inductive
settings. Our experiments show that we accomplish new scores on a widely used
knowledge graph benchmark. | cs.CL | Relations Prediction for Knowledge Graph Completion using Large Language Models |
2024-05-04T18:32:08Z | http://arxiv.org/pdf/2405.02732v1 | http://arxiv.org/abs/2405.02732v1 | Sneha Singhania, Simon Razniewski, Gerhard Weikum | 2024-05-04T18:32:08Z | cs.CL, cs.IR | Methods for relation extraction from text mostly focus on high precision, at
the cost of limited recall. High recall is crucial, though, to populate long
lists of object entities that stand in a specific relation with a given
subject. Cues for relevant objects can be spread across many passages in long
texts. This poses the challenge of extracting long lists from long texts. We
present the L3X method which tackles the problem in two stages: (1)
recall-oriented generation using a large language model (LLM) with judicious
techniques for retrieval augmentation, and (2) precision-oriented
scrutinization to validate or prune candidates. Our L3X method outperforms
LLM-only generations by a substantial margin. | cs.CL | Recall Them All: Retrieval-Augmented Language Models for Long Object List Extraction from Long Documents |
2024-05-04T17:10:00Z | http://arxiv.org/pdf/2405.02714v1 | http://arxiv.org/abs/2405.02714v1 | Xinran Zhao, Tong Chen, Sihao Chen, Hongming Zhang, Tongshuang Wu | 2024-05-04T17:10:00Z | cs.IR, cs.CL | The task of Information Retrieval (IR) requires a system to identify relevant
documents based on users' information needs. In real-world scenarios,
retrievers are expected to not only rely on the semantic relevance between the
documents and the queries but also recognize the nuanced intents or
perspectives behind a user query. For example, when asked to verify a claim, a
retrieval system is expected to identify evidence from both supporting vs.
contradicting perspectives, for the downstream system to make a fair judgment
call. In this work, we study whether retrievers can recognize and respond to
different perspectives of the queries -- beyond finding relevant documents for
a claim, can retrievers distinguish supporting vs. opposing documents? We
reform and extend six existing tasks to create a benchmark for retrieval, where
we have diverse perspectives described in free-form text, besides root, neutral
queries. We show that current retrievers covered in our experiments have
limited awareness of subtly different perspectives in queries and can also be
biased toward certain perspectives. Motivated by the observation, we further
explore the potential to leverage geometric features of retriever
representation space to improve the perspective awareness of retrievers in a
zero-shot manner. We demonstrate the efficiency and effectiveness of our
projection-based methods on the same set of tasks. Further analysis also shows
how perspective awareness improves performance on various downstream tasks,
with 4.2% higher accuracy on AmbigQA and 29.9% more correlation with designated
viewpoints on essay writing, compared to non-perspective-aware baselines. | cs.IR | Beyond Relevance: Evaluate and Improve Retrievers on Perspective Awareness |
2024-05-04T16:56:14Z | http://arxiv.org/pdf/2405.02712v1 | http://arxiv.org/abs/2405.02712v1 | Hanchong Zhang, Ruisheng Cao, Hongshen Xu, Lu Chen, Kai Yu | 2024-05-04T16:56:14Z | cs.CL | Recently, Large Language Models (LLMs) have been demonstrated to possess
impressive capabilities in a variety of domains and tasks. We investigate the
issue of prompt design in the multi-turn text-to-SQL task and attempt to
enhance the LLMs' reasoning capacity when generating SQL queries. In the
conversational context, the current SQL query can be modified from the
preceding SQL query with only a few operations due to the context dependency.
We introduce our method called CoE-SQL which can prompt LLMs to generate the
SQL query based on the previously generated SQL query with an edition chain. We
also conduct extensive ablation studies to determine the optimal configuration
of our approach. Our approach outperforms different in-context learning
baselines stably and achieves state-of-the-art performances on two benchmarks
SParC and CoSQL using LLMs, which is also competitive to the SOTA fine-tuned
models. | cs.CL | CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions |
2024-05-04T16:48:05Z | http://arxiv.org/pdf/2405.02710v1 | http://arxiv.org/abs/2405.02710v1 | Che Guan, Andrew Chin, Puya Vahabi | 2024-05-04T16:48:05Z | cs.CL | With the deluge of information delivered by the daily news cycle, there is a
growing need to effectively and efficiently summarize news feeds for quick
consumption. We leverage large language models (LLMs), with their advanced
learning and generative abilities as compared to conventional language models,
to generate concise and coherent summaries for news articles from the XSum
dataset. Our paper focuses on two key aspects of LLMs: Efficient in-context
Learning (ELearn) and Parameter Efficient Fine-tuning (EFit). Under ELearn, we
find that increasing the number of shots in prompts and utilizing simple
templates generally improve the quality of summaries. We also find that
utilizing relevant examples in few-shot learning for ELearn does not improve
model performance. In addition, we studied EFit using different methods and
demonstrate that fine-tuning the first layer of LLMs produces better outcomes
as compared to fine-tuning other layers or utilizing LoRA. We also find that
leveraging more relevant training samples using selective layers does not
result in better performance. By combining ELearn and EFit, we create a new
model (ELearnFit) that leverages the benefits of both few-shot learning and
fine-tuning and produces superior performance to either model alone. We also
use ELearnFit to highlight the trade-offs between prompting and fine-tuning,
especially for situations where only a limited number of annotated samples are
available. Ultimately, our research provides practical techniques to optimize
news summarization during the prompting and fine-tuning stages and enhances the
synthesis of news articles. | cs.CL | Enhancing News Summarization with ELearnFit through Efficient In-Context Learning and Efficient Fine-Tuning |
2024-05-04T15:40:17Z | http://arxiv.org/pdf/2405.06674v1 | http://arxiv.org/abs/2405.06674v1 | Xiaojun Chen, Tianle Wang, Tianhao Qiu, Jianbin Qin, Min Yang | 2024-05-04T15:40:17Z | cs.CL, cs.AI | Despite the success of large language models (LLMs) in Text-to-SQL tasks,
open-source LLMs encounter challenges in contextual understanding and response
coherence. To tackle these issues, we present \ours, a systematic methodology
tailored for Text-to-SQL with open-source LLMs. Our contributions include a
comprehensive evaluation of open-source LLMs in Text-to-SQL tasks, the
\openprompt strategy for effective question representation, and novel
strategies for supervised fine-tuning. We explore the benefits of
Chain-of-Thought in step-by-step inference and propose the \openexample method
for enhanced few-shot learning. Additionally, we introduce token-efficient
techniques, such as \textbf{Variable-length Open DB Schema}, \textbf{Target
Column Truncation}, and \textbf{Example Column Truncation}, addressing
challenges in large-scale databases. Our findings emphasize the need for
further investigation into the impact of supervised fine-tuning on contextual
learning capabilities. Remarkably, our method significantly improved Llama2-7B
from 2.54\% to 41.04\% and Code Llama-7B from 14.54\% to 48.24\% on the
BIRD-Dev dataset. Notably, the performance of Code Llama-7B surpassed GPT-4
(46.35\%) on the BIRD-Dev dataset. | cs.CL | Open-SQL Framework: Enhancing Text-to-SQL on Open-source Large Language Models |
2024-05-04T14:40:28Z | http://arxiv.org/pdf/2405.02677v1 | http://arxiv.org/abs/2405.02677v1 | Sebastián Concha Macías, Brian Keith Norambuena | 2024-05-04T14:40:28Z | cs.CL, cs.IR | Narratives serve as fundamental frameworks in our understanding of the world
and play a crucial role in collaborative sensemaking, providing a versatile
foundation for sensemaking. Framing is a subtle yet potent mechanism that
influences public perception through specific word choices, shaping
interpretations of reported news events. Despite the recognized importance of
narratives and framing, a significant gap exists in the literature with regard
to the explicit consideration of framing within the context of computational
extraction and representation. This article explores the capabilities of a
specific narrative extraction and representation approach -- narrative maps --
to capture framing information from news data. The research addresses two key
questions: (1) Does the narrative extraction method capture the framing
distribution of the data set? (2) Does it produce a representation with
consistent framing? Our results indicate that while the algorithm captures
framing distributions, achieving consistent framing across various starting and
ending events poses challenges. Our results highlight the potential of
narrative maps to provide users with insights into the intricate framing
dynamics within news narratives. However, we note that directly leveraging
framing information in the computational narrative extraction process remains
an open challenge. | cs.CL | Evaluating the Ability of Computationally Extracted Narrative Maps to Encode Media Framing |
2024-05-04T14:20:28Z | http://arxiv.org/pdf/2405.02673v1 | http://arxiv.org/abs/2405.02673v1 | Zhihao Wang, Longyue Wang, Jinsong Su, Junfeng Yao, Zhaopeng Tu | 2024-05-04T14:20:28Z | cs.CL | Token repetition is a typical form of multi-modal problem in fully
non-autoregressive translation (NAT). In this work, we revisit the multi-modal
problem in recently proposed NAT models. Our study reveals that these advanced
models have introduced other types of information redundancy errors, which
cannot be measured by the conventional metric - the continuous repetition
ratio. By manually annotating the NAT outputs, we identify two types of
information redundancy errors that correspond well to lexical and reordering
multi-modality problems. Since human annotation is time-consuming and
labor-intensive, we propose automatic metrics to evaluate the two types of
redundant errors. Our metrics allow future studies to evaluate new methods and
gain a more comprehensive understanding of their effectiveness. | cs.CL | On the Information Redundancy in Non-Autoregressive Translation |
2024-05-04T12:59:10Z | http://arxiv.org/pdf/2405.02659v2 | http://arxiv.org/abs/2405.02659v2 | Taolin Zhang, Dongyang Li, Qizhou Chen, Chengyu Wang, Longtao Huang, Hui Xue, Xiaofeng He, Jun Huang | 2024-06-25T09:24:01Z | cs.CL | Retrieval-augmented large language models (LLMs) leverage relevant content
retrieved by information retrieval systems to generate correct responses,
aiming to alleviate the hallucination problem. However, existing
retriever-responder methods typically append relevant documents to the prompt
of LLMs to perform text generation tasks without considering the interaction of
fine-grained structural semantics between the retrieved documents and the LLMs.
This issue is particularly important for accurate response generation as LLMs
tend to "lose in the middle" when dealing with input prompts augmented with
lengthy documents. In this work, we propose a new pipeline named "Reinforced
Retriever-Reorder-Responder" (R$^4$) to learn document orderings for
retrieval-augmented LLMs, thereby further enhancing their generation abilities
while the large numbers of parameters of LLMs remain frozen. The reordering
learning process is divided into two steps according to the quality of the
generated responses: document order adjustment and document representation
enhancement. Specifically, document order adjustment aims to organize retrieved
document orderings into beginning, middle, and end positions based on graph
attention learning, which maximizes the reinforced reward of response quality.
Document representation enhancement further refines the representations of
retrieved documents for responses of poor quality via document-level gradient
adversarial learning. Extensive experiments demonstrate that our proposed
pipeline achieves better factual question-answering performance on
knowledge-intensive tasks compared to strong baselines across various public
datasets. The source codes and trained models will be released upon paper
acceptance. | cs.CL | R4: Reinforced Retriever-Reorder-Responder for Retrieval-Augmented Large Language Models |
2024-05-04T12:29:00Z | http://arxiv.org/pdf/2405.02650v1 | http://arxiv.org/abs/2405.02650v1 | Maxim Ifergan, Renana Keydar, Omri Abend, Amit Pinchevski | 2024-05-04T12:29:00Z | cs.CL, cs.AI | The vast collection of Holocaust survivor testimonies presents invaluable
historical insights but poses challenges for manual analysis. This paper
leverages advanced Natural Language Processing (NLP) techniques to explore the
USC Shoah Foundation Holocaust testimony corpus. By treating testimonies as
structured question-and-answer sections, we apply topic modeling to identify
key themes. We experiment with BERTopic, which leverages recent advances in
language modeling technology. We align testimony sections into fixed parts,
revealing the evolution of topics across the corpus of testimonies. This
highlights both a common narrative schema and divergences between subgroups
based on age and gender. We introduce a novel method to identify testimonies
within groups that exhibit atypical topic distributions resembling those of
other groups. This study offers unique insights into the complex narratives of
Holocaust survivors, demonstrating the power of NLP to illuminate historical
discourse and identify potential deviations in survivor experiences. | cs.CL | Identifying Narrative Patterns and Outliers in Holocaust Testimonies Using Topic Modeling |
2024-05-04T11:22:16Z | http://arxiv.org/pdf/2405.02637v1 | http://arxiv.org/abs/2405.02637v1 | Mohammad Aliannejadi, Zahra Abbasiantaeb, Shubham Chatterjee, Jeffery Dalton, Leif Azzopardi | 2024-05-04T11:22:16Z | cs.IR, cs.AI, cs.CL | Conversational information seeking has evolved rapidly in the last few years
with the development of Large Language Models (LLMs), providing the basis for
interpreting and responding in a naturalistic manner to user requests. The
extended TREC Interactive Knowledge Assistance Track (iKAT) collection aims to
enable researchers to test and evaluate their Conversational Search Agents
(CSA). The collection contains a set of 36 personalized dialogues over 20
different topics each coupled with a Personal Text Knowledge Base (PTKB) that
defines the bespoke user personas. A total of 344 turns with approximately
26,000 passages are provided as assessments on relevance, as well as additional
assessments on generated responses over four key dimensions: relevance,
completeness, groundedness, and naturalness. The collection challenges CSA to
efficiently navigate diverse personal contexts, elicit pertinent persona
information, and employ context for relevant conversations. The integration of
a PTKB and the emphasis on decisional search tasks contribute to the uniqueness
of this test collection, making it an essential benchmark for advancing
research in conversational and interactive knowledge assistants. | cs.IR | TREC iKAT 2023: A Test Collection for Evaluating Conversational and Interactive Knowledge Assistants |
2024-05-04T08:04:39Z | http://arxiv.org/pdf/2405.02602v1 | http://arxiv.org/abs/2405.02602v1 | Julia Evans, Sameer Sadruddin, Jennifer D'Souza | 2024-05-04T08:04:39Z | cs.CL, cs.AI, cs.IT, math.IT | In this study, we address one of the challenges of developing NER models for
scholarly domains, namely the scarcity of suitable labeled data. We experiment
with an approach using predictions from a fine-tuned LLM model to aid
non-domain experts in annotating scientific entities within astronomy
literature, with the goal of uncovering whether such a collaborative process
can approximate domain expertise. Our results reveal moderate agreement between
a domain expert and the LLM-assisted non-experts, as well as fair agreement
between the domain expert and the LLM model's predictions. In an additional
experiment, we compare the performance of finetuned and default LLMs on this
task. We have also introduced a specialized scientific entity annotation scheme
for astronomy, validated by a domain expert. Our approach adopts a scholarly
research contribution-centric perspective, focusing exclusively on scientific
entities relevant to the research theme. The resultant dataset, containing
5,000 annotated astronomy article titles, is made publicly available. | cs.CL | Astro-NER -- Astronomy Named Entity Recognition: Is GPT a Good Domain Expert Annotator? |
2024-05-04T07:44:18Z | http://arxiv.org/pdf/2405.02596v1 | http://arxiv.org/abs/2405.02596v1 | Jing Xu, Jingzhao Zhang | 2024-05-04T07:44:18Z | cs.LG, cs.AI, cs.CL | Fine-tuning large language models (LLM) can be costly. Parameter-efficient
fine-tuning (PEFT) addresses the problems by training a fraction of the
parameters, whose success reveals the expressiveness and flexibility of
pretrained models. This paper studies the limit of PEFT, by further simplifying
its design and reducing the number of trainable parameters beyond standard
setups. To this end, we use Random Masking to fine-tune the pretrained model.
Despite its simplicity, we show that Random Masking is surprisingly effective:
with a larger-than-expected learning rate, Random Masking can match the
performance of standard PEFT algorithms such as LoRA on various tasks, using
fewer trainable parameters. We provide both empirical and theoretical
explorations into the success of Random Masking. We show that masking induces a
flatter loss landscape and more distant solutions, which allows for and
necessitates large learning rates. | cs.LG | Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning |
2024-05-04T06:06:34Z | http://arxiv.org/pdf/2405.02578v1 | http://arxiv.org/abs/2405.02578v1 | Maryam Al Ali, Hanan Aldarmaki | 2024-05-04T06:06:34Z | cs.CL | This paper introduces Mixat: a dataset of Emirati speech code-mixed with
English. Mixat was developed to address the shortcomings of current speech
recognition resources when applied to Emirati speech, and in particular, to
bilignual Emirati speakers who often mix and switch between their local dialect
and English. The data set consists of 15 hours of speech derived from two
public podcasts featuring native Emirati speakers, one of which is in the form
of conversations between the host and a guest. Therefore, the collection
contains examples of Emirati-English code-switching in both formal and natural
conversational contexts. In this paper, we describe the process of data
collection and annotation, and describe some of the features and statistics of
the resulting data set. In addition, we evaluate the performance of pre-trained
Arabic and multi-lingual ASR systems on our dataset, demonstrating the
shortcomings of existing models on this low-resource dialectal Arabic, and the
additional challenge of recognizing code-switching in ASR. The dataset will be
made publicly available for research use. | cs.CL | Mixat: A Data Set of Bilingual Emirati-English Speech |
2024-05-04T05:24:19Z | http://arxiv.org/pdf/2405.02573v1 | http://arxiv.org/abs/2405.02573v1 | Hieu Ngo Trung, Duong Tran Ham, Tin Huynh, Kiem Hoang | 2024-05-04T05:24:19Z | cs.CL | Recently, many studies have shown the efficiency of using Bidirectional
Encoder Representations from Transformers (BERT) in various Natural Language
Processing (NLP) tasks. Specifically, English spelling correction task that
uses Encoder-Decoder architecture and takes advantage of BERT has achieved
state-of-the-art result. However, to our knowledge, there is no implementation
in Vietnamese yet. Therefore, in this study, a combination of Transformer
architecture (state-of-the-art for Encoder-Decoder model) and BERT was proposed
to deal with Vietnamese spelling correction. The experiment results have shown
that our model outperforms other approaches as well as the Google Docs Spell
Checking tool, achieves an 86.24 BLEU score on this task. | cs.CL | A Combination of BERT and Transformer for Vietnamese Spelling Correction |
2024-05-04T04:16:07Z | http://arxiv.org/pdf/2405.02559v1 | http://arxiv.org/abs/2405.02559v1 | Thomas Yu Chow Tam, Sonish Sivarajkumar, Sumit Kapoor, Alisa V Stolyar, Katelyn Polanska, Karleigh R McCarthy, Hunter Osterhoudt, Xizhi Wu, Shyam Visweswaran, Sunyang Fu, Piyush Mathur, Giovanni E. Cacciamani, Cong Sun, Yifan Peng, Yanshan Wang | 2024-05-04T04:16:07Z | cs.CL, cs.AI | As generative artificial intelligence (AI), particularly Large Language
Models (LLMs), continues to permeate healthcare, it remains crucial to
supplement traditional automated evaluations with human expert evaluation.
Understanding and evaluating the generated texts is vital for ensuring safety,
reliability, and effectiveness. However, the cumbersome, time-consuming, and
non-standardized nature of human evaluation presents significant obstacles to
the widespread adoption of LLMs in practice. This study reviews existing
literature on human evaluation methodologies for LLMs within healthcare. We
highlight a notable need for a standardized and consistent human evaluation
approach. Our extensive literature search, adhering to the Preferred Reporting
Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, spans
publications from January 2018 to February 2024. This review provides a
comprehensive overview of the human evaluation approaches used in diverse
healthcare applications.This analysis examines the human evaluation of LLMs
across various medical specialties, addressing factors such as evaluation
dimensions, sample types, and sizes, the selection and recruitment of
evaluators, frameworks and metrics, the evaluation process, and statistical
analysis of the results. Drawing from diverse evaluation strategies highlighted
in these studies, we propose a comprehensive and practical framework for human
evaluation of generative LLMs, named QUEST: Quality of Information,
Understanding and Reasoning, Expression Style and Persona, Safety and Harm, and
Trust and Confidence. This framework aims to improve the reliability,
generalizability, and applicability of human evaluation of generative LLMs in
different healthcare applications by defining clear evaluation dimensions and
offering detailed guidelines. | cs.CL | A Literature Review and Framework for Human Evaluation of Generative Large Language Models in Healthcare |
2024-05-04T04:12:18Z | http://arxiv.org/pdf/2405.06673v2 | http://arxiv.org/abs/2405.06673v2 | Gyubok Lee, Sunjun Kweon, Seongsu Bae, Edward Choi | 2024-05-23T17:25:21Z | cs.CL, cs.AI | Electronic Health Records (EHRs) are relational databases that store the
entire medical histories of patients within hospitals. They record numerous
aspects of patients' medical care, from hospital admission and diagnosis to
treatment and discharge. While EHRs are vital sources of clinical data,
exploring them beyond a predefined set of queries requires skills in query
languages like SQL. To make information retrieval more accessible, one strategy
is to build a question-answering system, possibly leveraging text-to-SQL models
that can automatically translate natural language questions into corresponding
SQL queries and use these queries to retrieve the answers. The EHRSQL 2024
shared task aims to advance and promote research in developing a
question-answering system for EHRs using text-to-SQL modeling, capable of
reliably providing requested answers to various healthcare professionals to
improve their clinical work processes and satisfy their needs. Among more than
100 participants who applied to the shared task, eight teams were formed and
completed the entire shared task requirement and demonstrated a wide range of
methods to effectively solve this task. In this paper, we describe the task of
reliable text-to-SQL modeling, the dataset, and the methods and results of the
participants. We hope this shared task will spur further research and insights
into developing reliable question-answering systems for EHRs. | cs.CL | Overview of the EHRSQL 2024 Shared Task on Reliable Text-to-SQL Modeling on Electronic Health Records |
2024-05-03T23:04:52Z | http://arxiv.org/pdf/2405.02517v1 | http://arxiv.org/abs/2405.02517v1 | Alvin Po-Chun Chen, Ray Groshan, Sean von Bayern | 2024-05-03T23:04:52Z | cs.CL | Extensive research exists on the performance of large language models on
logic-based tasks, whereas relatively little has been done on their ability to
generate creative solutions on lateral thinking tasks. The BrainTeaser shared
task tests lateral thinking and uses adversarial datasets to prevent
memorization, resulting in poor performance for out-of-the-box models. We
propose a system for iterative, chain-of-thought prompt engineering which
optimizes prompts using human evaluation. Using this shared task, we
demonstrate our system's ability to significantly improve model performance by
optimizing prompts and evaluate the input dataset. | cs.CL | Mothman at SemEval-2024 Task 9: An Iterative System for Chain-of-Thought Prompt Optimization |
2024-05-03T22:17:22Z | http://arxiv.org/pdf/2405.02501v2 | http://arxiv.org/abs/2405.02501v2 | Hyeong Kyu Choi, Yixuan Li | 2024-05-14T05:53:07Z | cs.CL, cs.AI | Large Language Models (LLMs) are trained on massive text corpora, which are
encoded with diverse personality traits. This triggers an interesting goal of
eliciting a desired personality trait from the LLM, and probing its behavioral
preferences. Accordingly, we formalize the persona elicitation task, aiming to
customize LLM behaviors to align with a target persona. We present Persona
In-Context Learning (PICLe), a novel persona elicitation framework grounded in
Bayesian inference. At the core, PICLe introduces a new ICL example selection
criterion based on likelihood ratio, which is designed to optimally guide the
model in eliciting a specific target persona. We demonstrate the effectiveness
of PICLe through extensive comparisons against baseline methods across three
contemporary LLMs. Code is available at
https://github.com/deeplearning-wisc/picle. | cs.CL | PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learning |
2024-05-03T20:20:15Z | http://arxiv.org/pdf/2405.02472v1 | http://arxiv.org/abs/2405.02472v1 | Michael Burnham | 2024-05-03T20:20:15Z | cs.CL | This paper introduces "Semantic Scaling," a novel method for ideal point
estimation from text. I leverage large language models to classify documents
based on their expressed stances and extract survey-like data. I then use item
response theory to scale subjects from these data. Semantic Scaling
significantly improves on existing text-based scaling methods, and allows
researchers to explicitly define the ideological dimensions they measure. This
represents the first scaling approach that allows such flexibility outside of
survey instruments and opens new avenues of inquiry for populations difficult
to survey. Additionally, it works with documents of varying length, and
produces valid estimates of both mass and elite ideology. I demonstrate that
the method can differentiate between policy preferences and in-group/out-group
affect. Among the public, Semantic Scaling out-preforms Tweetscores according
to human judgement; in Congress, it recaptures the first dimension DW-NOMINATE
while allowing for greater flexibility in resolving construct validity
challenges. | cs.CL | Semantic Scaling: Bayesian Ideal Point Estimates with Large Language Models |
2024-05-03T19:37:37Z | http://arxiv.org/pdf/2405.02454v1 | http://arxiv.org/abs/2405.02454v1 | Michael Burnham | 2024-05-03T19:37:37Z | cs.CL, cs.AI | Sentiment analysis is one of the most widely used techniques in text
analysis. Recent advancements with Large Language Models have made it more
accurate and accessible than ever, allowing researchers to classify text with
only a plain English prompt. However, "sentiment" entails a wide variety of
concepts depending on the domain and tools used. It has been used to mean
emotion, opinions, market movements, or simply a general ``good-bad''
dimension. This raises a question: What exactly are language models doing when
prompted to label documents by sentiment? This paper first overviews how
sentiment is defined across different contexts, highlighting that it is a
confounded measurement construct in that it entails multiple variables, such as
emotional valence and opinion, without disentangling them. I then test three
language models across two data sets with prompts requesting sentiment,
valence, and stance classification. I find that sentiment labels most strongly
correlate with valence labels. I further find that classification improves when
researchers more precisely specify their dimension of interest rather than
using the less well-defined concept of sentiment. I conclude by encouraging
researchers to move beyond "sentiment" when feasible and use a more precise
measurement construct. | cs.CL | What is Sentiment Meant to Mean to Language Models? |
2024-05-03T19:24:41Z | http://arxiv.org/pdf/2406.02566v1 | http://arxiv.org/abs/2406.02566v1 | Ognjen Kundacina, Vladimir Vincan, Dragisa Miskovic | 2024-05-03T19:24:41Z | eess.AS, cs.AI, cs.CL, cs.LG | Emphasizing a data-centric AI approach, this paper introduces a novel
two-stage active learning (AL) pipeline for automatic speech recognition (ASR),
combining unsupervised and supervised AL methods. The first stage utilizes
unsupervised AL by using x-vectors clustering for diverse sample selection from
unlabeled speech data, thus establishing a robust initial dataset for the
subsequent supervised AL. The second stage incorporates a supervised AL
strategy, with a batch AL method specifically developed for ASR, aimed at
selecting diverse and informative batches of samples. Here, sample diversity is
also achieved using x-vectors clustering, while the most informative samples
are identified using a Bayesian AL method tailored for ASR with an adaptation
of Monte Carlo dropout to approximate Bayesian inference. This approach enables
precise uncertainty estimation, thereby enhancing ASR model training with
significantly reduced data requirements. Our method has shown superior
performance compared to competing methods on homogeneous, heterogeneous, and
OOD test sets, demonstrating that strategic sample selection and innovative
Bayesian modeling can substantially optimize both labeling effort and data
utilization in deep learning-based ASR applications. | eess.AS | Combining X-Vectors and Bayesian Batch Active Learning: Two-Stage Active Learning Pipeline for Speech Recognition |
2024-05-03T18:51:19Z | http://arxiv.org/pdf/2405.02429v1 | http://arxiv.org/abs/2405.02429v1 | Yaoyiran Li, Xiang Zhai, Moustafa Alzantot, Keyi Yu, Ivan Vulić, Anna Korhonen, Mohamed Hammad | 2024-05-03T18:51:19Z | cs.IR, cs.AI, cs.CL, cs.LG | Traditional recommender systems such as matrix factorization methods rely on
learning a shared dense embedding space to represent both items and user
preferences. Sequence models such as RNN, GRUs, and, recently, Transformers
have also excelled in the task of sequential recommendation. This task requires
understanding the sequential structure present in users' historical
interactions to predict the next item they may like. Building upon the success
of Large Language Models (LLMs) in a variety of tasks, researchers have
recently explored using LLMs that are pretrained on vast corpora of text for
sequential recommendation. To use LLMs in sequential recommendations, both the
history of user interactions and the model's prediction of the next item are
expressed in text form. We propose CALRec, a two-stage LLM finetuning framework
that finetunes a pretrained LLM in a two-tower fashion using a mixture of two
contrastive losses and a language modeling loss: the LLM is first finetuned on
a data mixture from multiple domains followed by another round of target domain
finetuning. Our model significantly outperforms many state-of-the-art baselines
(+37% in Recall@1 and +24% in NDCG@10) and systematic ablation studies reveal
that (i) both stages of finetuning are crucial, and, when combined, we achieve
improved performance, and (ii) contrastive alignment is effective among the
target domains explored in our experiments. | cs.IR | CALRec: Contrastive Alignment of Generative LLMs For Sequential Recommendation |
2024-05-03T18:34:37Z | http://arxiv.org/pdf/2405.02421v1 | http://arxiv.org/abs/2405.02421v1 | Jingcheng Niu, Andrew Liu, Zining Zhu, Gerald Penn | 2024-05-03T18:34:37Z | cs.CL | We reassess the Knowledge Neuron (KN) Thesis: an interpretation of the
mechanism underlying the ability of large language models to recall facts from
a training corpus. This nascent thesis proposes that facts are recalled from
the training corpus through the MLP weights in a manner resembling key-value
memory, implying in effect that "knowledge" is stored in the network.
Furthermore, by modifying the MLP modules, one can control the language model's
generation of factual information. The plausibility of the KN thesis has been
demonstrated by the success of KN-inspired model editing methods (Dai et al.,
2022; Meng et al., 2022).
We find that this thesis is, at best, an oversimplification. Not only have we
found that we can edit the expression of certain linguistic phenomena using the
same model editing methods but, through a more comprehensive evaluation, we
have found that the KN thesis does not adequately explain the process of
factual expression. While it is possible to argue that the MLP weights store
complex patterns that are interpretable both syntactically and semantically,
these patterns do not constitute "knowledge." To gain a more comprehensive
understanding of the knowledge representation process, we must look beyond the
MLP weights and explore recent models' complex layer structures and attention
mechanisms. | cs.CL | What does the Knowledge Neuron Thesis Have to do with Knowledge? |
2024-05-03T18:12:39Z | http://arxiv.org/pdf/2405.02411v1 | http://arxiv.org/abs/2405.02411v1 | Diyi Yang, Dirk Hovy, David Jurgens, Barbara Plank | 2024-05-03T18:12:39Z | cs.CL | Language technologies have made enormous progress, especially with the
introduction of large language models (LLMs). On traditional tasks such as
machine translation and sentiment analysis, these models perform at near-human
level. These advances can, however, exacerbate a variety of issues that models
have traditionally struggled with, such as bias, evaluation, and risks. In this
position paper, we argue that many of these issues share a common core: a lack
of awareness of the factors, context, and implications of the social
environment in which NLP operates, which we call social awareness. While NLP is
getting better at solving the formal linguistic aspects, limited progress has
been made in adding the social awareness required for language applications to
work in all situations for all users. Integrating social awareness into NLP
models will make applications more natural, helpful, and safe, and will open up
new possibilities. Thus we argue that substantial challenges remain for NLP to
develop social awareness and that we are just at the beginning of a new era for
the field. | cs.CL | The Call for Socially Aware Language Technologies |
2024-05-03T17:59:55Z | http://arxiv.org/pdf/2405.02287v1 | http://arxiv.org/abs/2405.02287v1 | Piotr Padlewski, Max Bain, Matthew Henderson, Zhongkai Zhu, Nishant Relan, Hai Pham, Donovan Ong, Kaloyan Aleksiev, Aitor Ormazabal, Samuel Phua, Ethan Yeo, Eugenie Lamprecht, Qi Liu, Yuqi Wang, Eric Chen, Deyu Fu, Lei Li, Che Zheng, Cyprien de Masson d'Autume, Dani Yogatama, Mikel Artetxe, Yi Tay | 2024-05-03T17:59:55Z | cs.CL, cs.AI, cs.CV | We introduce Vibe-Eval: a new open benchmark and framework for evaluating
multimodal chat models. Vibe-Eval consists of 269 visual understanding prompts,
including 100 of hard difficulty, complete with gold-standard responses
authored by experts. Vibe-Eval is open-ended and challenging with dual
objectives: (i) vibe checking multimodal chat models for day-to-day tasks and
(ii) rigorously testing and probing the capabilities of present frontier
models. Notably, our hard set contains >50% questions that all frontier models
answer incorrectly. We explore the nuances of designing, evaluating, and
ranking models on ultra challenging prompts. We also discuss trade-offs between
human and automatic evaluation, and show that automatic model evaluation using
Reka Core roughly correlates to human judgment. We offer free API access for
the purpose of lightweight evaluation and plan to conduct formal human
evaluations for public models that perform well on the Vibe-Eval's automatic
scores. We release the evaluation code and data, see
https://github.com/reka-ai/reka-vibe-eval | cs.CL | Vibe-Eval: A hard evaluation suite for measuring progress of multimodal language models |
2024-05-03T17:34:57Z | http://arxiv.org/pdf/2405.02267v1 | http://arxiv.org/abs/2405.02267v1 | Aaron Klein, Jacek Golebiowski, Xingchen Ma, Valerio Perrone, Cedric Archambeau | 2024-05-03T17:34:57Z | cs.LG, cs.CL | Pre-trained language models (PLM), for example BERT or RoBERTa, mark the
state-of-the-art for natural language understanding task when fine-tuned on
labeled data. However, their large size poses challenges in deploying them for
inference in real-world applications, due to significant GPU memory
requirements and high inference latency. This paper explores neural
architecture search (NAS) for structural pruning to find sub-parts of the
fine-tuned network that optimally trade-off efficiency, for example in terms of
model size or latency, and generalization performance. We also show how we can
utilize more recently developed two-stage weight-sharing NAS approaches in this
setting to accelerate the search process. Unlike traditional pruning methods
with fixed thresholds, we propose to adopt a multi-objective approach that
identifies the Pareto optimal set of sub-networks, allowing for a more flexible
and automated compression process. | cs.LG | Structural Pruning of Pre-trained Language Models via Neural Architecture Search |
2024-05-03T16:41:36Z | http://arxiv.org/pdf/2405.06671v2 | http://arxiv.org/abs/2405.06671v2 | Subhendu Khatuya, Rajdeep Mukherjee, Akash Ghosh, Manjunath Hegde, Koustuv Dasgupta, Niloy Ganguly, Saptarshi Ghosh, Pawan Goyal | 2024-05-15T14:43:23Z | cs.CL, cs.CE, cs.LG | We study the problem of automatically annotating relevant numerals (GAAP
metrics) occurring in the financial documents with their corresponding XBRL
tags. Different from prior works, we investigate the feasibility of solving
this extreme classification problem using a generative paradigm through
instruction tuning of Large Language Models (LLMs). To this end, we leverage
metric metadata information to frame our target outputs while proposing a
parameter efficient solution for the task using LoRA. We perform experiments on
two recently released financial numeric labeling datasets. Our proposed model,
FLAN-FinXC, achieves new state-of-the-art performances on both the datasets,
outperforming several strong baselines. We explain the better scores of our
proposed model by demonstrating its capability for zero-shot as well as the
least frequently occurring tags. Also, even when we fail to predict the XBRL
tags correctly, our generated output has substantial overlap with the
ground-truth in majority of the cases. | cs.CL | Parameter-Efficient Instruction Tuning of Large Language Models For Extreme Financial Numeral Labelling |
2024-05-03T16:38:51Z | http://arxiv.org/pdf/2405.02228v2 | http://arxiv.org/abs/2405.02228v2 | Deepa Tilwani, Yash Saxena, Ali Mohammadi, Edward Raff, Amit Sheth, Srinivasan Parthasarathy, Manas Gaur | 2024-05-09T00:23:16Z | cs.CL, cs.AI, cs.IR | Automatic citation generation for sentences in a document or report is
paramount for intelligence analysts, cybersecurity, news agencies, and
education personnel. In this research, we investigate whether large language
models (LLMs) are capable of generating references based on two forms of
sentence queries: (a) Direct Queries, LLMs are asked to provide author names of
the given research article, and (b) Indirect Queries, LLMs are asked to provide
the title of a mentioned article when given a sentence from a different
article. To demonstrate where LLM stands in this task, we introduce a large
dataset called REASONS comprising abstracts of the 12 most popular domains of
scientific research on arXiv. From around 20K research articles, we make the
following deductions on public and proprietary LLMs: (a) State-of-the-art,
often called anthropomorphic GPT-4 and GPT-3.5, suffers from high pass
percentage (PP) to minimize the hallucination rate (HR). When tested with
Perplexity.ai (7B), they unexpectedly made more errors; (b) Augmenting relevant
metadata lowered the PP and gave the lowest HR; (c) Advance retrieval-augmented
generation (RAG) using Mistral demonstrates consistent and robust citation
support on indirect queries and matched performance to GPT-3.5 and GPT-4. The
HR across all domains and models decreased by an average of 41.93%, and the PP
was reduced to 0% in most cases. In terms of generation quality, the average F1
Score and BLEU were 68.09% and 57.51%, respectively; (d) Testing with
adversarial samples showed that LLMs, including the Advance RAG Mistral,
struggle to understand context, but the extent of this issue was small in
Mistral and GPT-4-Preview. Our study contributes valuable insights into the
reliability of RAG for automated citation generation tasks. | cs.CL | REASONS: A benchmark for REtrieval and Automated citationS Of scieNtific Sentences using Public and Proprietary LLMs |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.