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
|
|---|---|---|---|---|---|---|---|---|
2024-01-24T06:55:55Z | http://arxiv.org/pdf/2401.13260v2 | http://arxiv.org/abs/2401.13260v2 | Jiajun He, Xiaohan Shi, Xingfeng Li, Tomoki Toda | 2024-05-28T16:47:15Z | cs.CL, cs.MM, cs.SD, eess.AS | The prevalent approach in speech emotion recognition (SER) involves
integrating both audio and textual information to comprehensively identify the
speaker's emotion, with the text generally obtained through automatic speech
recognition (ASR). An essential issue of this approach is that ASR errors from
the text modality can worsen the performance of SER. Previous studies have
proposed using an auxiliary ASR error detection task to adaptively assign
weights of each word in ASR hypotheses. However, this approach has limited
improvement potential because it does not address the coherence of semantic
information in the text. Additionally, the inherent heterogeneity of different
modalities leads to distribution gaps between their representations, making
their fusion challenging. Therefore, in this paper, we incorporate two
auxiliary tasks, ASR error detection (AED) and ASR error correction (AEC), to
enhance the semantic coherence of ASR text, and further introduce a novel
multi-modal fusion (MF) method to learn shared representations across
modalities. We refer to our method as MF-AED-AEC. Experimental results indicate
that MF-AED-AEC significantly outperforms the baseline model by a margin of
4.1\%. | cs.CL | MF-AED-AEC: Speech Emotion Recognition by Leveraging Multimodal Fusion, Asr Error Detection, and Asr Error Correction |
2024-01-24T06:50:20Z | http://arxiv.org/pdf/2401.13256v1 | http://arxiv.org/abs/2401.13256v1 | Hongru Wang, Wenyu Huang, Yang Deng, Rui Wang, Zezhong Wang, Yufei Wang, Fei Mi, Jeff Z. Pan, Kam-Fai Wong | 2024-01-24T06:50:20Z | cs.CL, cs.AI | Large Language Models (LLMs) has shown exceptional capabilities in many
natual language understanding and generation tasks. However, the
personalization issue still remains a much-coveted property, especially when it
comes to the multiple sources involved in the dialogue system. To better plan
and incorporate the use of multiple sources in generating personalized
response, we firstly decompose it into three sub-tasks: Knowledge Source
Selection, Knowledge Retrieval, and Response Generation. We then propose a
novel Unified Multi-Source Retrieval-Augmented Generation system (UniMS-RAG)
Specifically, we unify these three sub-tasks with different formulations into
the same sequence-to-sequence paradigm during the training, to adaptively
retrieve evidences and evaluate the relevance on-demand using special tokens,
called acting tokens and evaluation tokens. Enabling language models to
generate acting tokens facilitates interaction with various knowledge sources,
allowing them to adapt their behavior to diverse task requirements. Meanwhile,
evaluation tokens gauge the relevance score between the dialogue context and
the retrieved evidence. In addition, we carefully design a self-refinement
mechanism to iteratively refine the generated response considering 1) the
consistency scores between the generated response and retrieved evidence; and
2) the relevance scores. Experiments on two personalized datasets (DuLeMon and
KBP) show that UniMS-RAG achieves state-of-the-art performance on the knowledge
source selection and response generation task with itself as a retriever in a
unified manner. Extensive analyses and discussions are provided for shedding
some new perspectives for personalized dialogue systems. | cs.CL | UniMS-RAG: A Unified Multi-source Retrieval-Augmented Generation for Personalized Dialogue Systems |
2024-01-24T06:10:51Z | http://arxiv.org/pdf/2401.13246v3 | http://arxiv.org/abs/2401.13246v3 | Guoxin Chen, Kexin Tang, Chao Yang, Fuying Ye, Yu Qiao, Yiming Qian | 2024-06-04T06:14:56Z | cs.CL | Elucidating the reasoning process with structured explanations from question
to answer is crucial, as it significantly enhances the interpretability,
traceability, and trustworthiness of question-answering (QA) systems. However,
structured explanations demand models to perform intricately structured
reasoning, which poses great challenges. Most existing methods focus on
single-step reasoning through supervised learning, ignoring logical
dependencies between steps. Moreover, existing reinforcement learning (RL)
based methods overlook the structured relationships, underutilizing the
potential of RL in structured reasoning. In this paper, we propose SEER, a
novel method that maximizes a structure-based return to facilitate structured
reasoning and explanation. Our proposed structure-based return precisely
describes the hierarchical and branching structure inherent in structured
reasoning, effectively capturing the intricate relationships between different
reasoning steps. In addition, we introduce a fine-grained reward function to
meticulously delineate diverse reasoning steps. Extensive experiments show that
SEER significantly outperforms state-of-the-art methods, achieving an absolute
improvement of 6.9% over RL-based methods on EntailmentBank, a 4.4% average
improvement on STREET benchmark, and exhibiting outstanding efficiency and
cross-dataset generalization performance. Our code is available at
https://github.com/Chen-GX/SEER. | cs.CL | SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learning |
2024-01-24T04:57:32Z | http://arxiv.org/pdf/2401.13229v1 | http://arxiv.org/abs/2401.13229v1 | Alexandre Alcoforado, Thomas Palmeira Ferraz, Lucas Hideki Okamura, Israel Campos Fama, Arnold Moya Lavado, Bárbara Dias Bueno, Bruno Veloso, Anna Helena Reali Costa | 2024-01-24T04:57:32Z | cs.CL, cs.AI, cs.LG | A major challenge in Natural Language Processing is obtaining annotated data
for supervised learning. An option is the use of crowdsourcing platforms for
data annotation. However, crowdsourcing introduces issues related to the
annotator's experience, consistency, and biases. An alternative is to use
zero-shot methods, which in turn have limitations compared to their few-shot or
fully supervised counterparts. Recent advancements driven by large language
models show potential, but struggle to adapt to specialized domains with
severely limited data. The most common approaches therefore involve the human
itself randomly annotating a set of datapoints to build initial datasets. But
randomly sampling data to be annotated is often inefficient as it ignores the
characteristics of the data and the specific needs of the model. The situation
worsens when working with imbalanced datasets, as random sampling tends to
heavily bias towards the majority classes, leading to excessive annotated data.
To address these issues, this paper contributes an automatic and informed data
selection architecture to build a small dataset for few-shot learning. Our
proposal minimizes the quantity and maximizes diversity of data selected for
human annotation, while improving model performance. | cs.CL | From Random to Informed Data Selection: A Diversity-Based Approach to Optimize Human Annotation and Few-Shot Learning |
2024-01-24T04:50:16Z | http://arxiv.org/pdf/2401.13227v3 | http://arxiv.org/abs/2401.13227v3 | Baolong Bi, Shenghua Liu, Yiwei Wang, Lingrui Mei, Xueqi Cheng | 2024-02-20T03:53:06Z | cs.CL, cs.AI, cs.LG, cs.SI | Exploring the application of large language models (LLMs) to graph learning
is a emerging endeavor. However, the vast amount of information inherent in
large graphs poses significant challenges to this process. This work focuses on
the link prediction task and introduces $\textbf{LPNL}$ (Link Prediction via
Natural Language), a framework based on large language models designed for
scalable link prediction on large-scale heterogeneous graphs. We design novel
prompts for link prediction that articulate graph details in natural language.
We propose a two-stage sampling pipeline to extract crucial information from
the graphs, and a divide-and-conquer strategy to control the input tokens
within predefined limits, addressing the challenge of overwhelming information.
We fine-tune a T5 model based on our self-supervised learning designed for link
prediction. Extensive experimental results demonstrate that LPNL outperforms
multiple advanced baselines in link prediction tasks on large-scale graphs. | cs.CL | LPNL: Scalable Link Prediction with Large Language Models |
2024-01-24T04:44:42Z | http://arxiv.org/pdf/2402.01696v2 | http://arxiv.org/abs/2402.01696v2 | Vidit Jain, Mukund Rungta, Yuchen Zhuang, Yue Yu, Zeyu Wang, Mu Gao, Jeffrey Skolnick, Chao Zhang | 2024-02-22T03:16:46Z | cs.CL, cs.LG | Hierarchical text classification (HTC) is a complex subtask under multi-label
text classification, characterized by a hierarchical label taxonomy and data
imbalance. The best-performing models aim to learn a static representation by
combining document and hierarchical label information. However, the relevance
of document sections can vary based on the hierarchy level, necessitating a
dynamic document representation. To address this, we propose HiGen, a
text-generation-based framework utilizing language models to encode dynamic
text representations. We introduce a level-guided loss function to capture the
relationship between text and label name semantics. Our approach incorporates a
task-specific pretraining strategy, adapting the language model to in-domain
knowledge and significantly enhancing performance for classes with limited
examples. Furthermore, we present a new and valuable dataset called ENZYME,
designed for HTC, which comprises articles from PubMed with the goal of
predicting Enzyme Commission (EC) numbers. Through extensive experiments on the
ENZYME dataset and the widely recognized WOS and NYT datasets, our methodology
demonstrates superior performance, surpassing existing approaches while
efficiently handling data and mitigating class imbalance. The data and code
will be released publicly. | cs.CL | HiGen: Hierarchy-Aware Sequence Generation for Hierarchical Text Classification |
2024-01-24T04:28:50Z | http://arxiv.org/pdf/2401.13223v2 | http://arxiv.org/abs/2401.13223v2 | Fengbin Zhu, Ziyang Liu, Fuli Feng, Chao Wang, Moxin Li, Tat-Seng Chua | 2024-02-22T13:36:56Z | cs.CL, cs.AI | In this work, we address question answering (QA) over a hybrid of tabular and
textual data that are very common content on the Web (e.g. SEC filings), where
discrete reasoning capabilities are often required. Recently, large language
models (LLMs) like GPT-4 have demonstrated strong multi-step reasoning
capabilities. We then consider harnessing the amazing power of LLMs to solve
our task. We abstract a Step-wise Pipeline for tabular and textual QA, which
consists of three key steps, including Extractor, Reasoner and Executor, and
initially design an instruction to instantiate the pipeline and validate that
GPT-4 outperforms all existing methods. However, utilizing an online LLM like
GPT-4 holds various challenges in terms of cost, latency, and data security
risk, which motivates us to specialize smaller LLMs in this task. We develop a
TAT-LLM language model by fine-tuning LLaMA 2 with the training data generated
automatically from existing expert-annotated datasets following the Step-wise
Pipeline. The experimental results have verified that our TAT-LLM model can
outperform all baseline models, including the previous best fine-tuned models
and very large-scale LLMs like GPT-4 on FinQA, TAT-QA and TAT-DQA benchmarks. | cs.CL | TAT-LLM: A Specialized Language Model for Discrete Reasoning over Tabular and Textual Data |
2024-01-24T04:13:28Z | http://arxiv.org/pdf/2401.13218v1 | http://arxiv.org/abs/2401.13218v1 | Xinliang Frederick Zhang, Carter Blum, Temma Choji, Shalin Shah, Alakananda Vempala | 2024-01-24T04:13:28Z | cs.CL | Structural extraction of events within discourse is critical since it avails
a deeper understanding of communication patterns and behavior trends. Event
argument extraction (EAE), at the core of event-centric understanding, is the
task of identifying role-specific text spans (i.e., arguments) for a given
event. Document-level EAE (DocEAE) focuses on arguments that are scattered
across an entire document. In this work, we explore the capabilities of open
source Large Language Models (LLMs), i.e., Flan-UL2, for the DocEAE task. To
this end, we propose ULTRA, a hierarchical framework that extracts event
arguments more cost-effectively -- the method needs as few as 50 annotations
and doesn't require hitting costly API endpoints. Further, it alleviates the
positional bias issue intrinsic to LLMs. ULTRA first sequentially reads text
chunks of a document to generate a candidate argument set, upon which ULTRA
learns to drop non-pertinent candidates through self-refinement. We further
introduce LEAFER to address the challenge LLMs face in locating the exact
boundary of an argument span. ULTRA outperforms strong baselines, which include
strong supervised models and ChatGPT, by 9.8% when evaluated by the exact match
(EM) metric. | cs.CL | ULTRA: Unleash LLMs' Potential for Event Argument Extraction through Hierarchical Modeling and Pair-wise Refinement |
2024-01-24T03:53:28Z | http://arxiv.org/pdf/2401.16212v1 | http://arxiv.org/abs/2401.16212v1 | Lauren Martin, Nick Whitehouse, Stephanie Yiu, Lizzie Catterson, Rivindu Perera | 2024-01-24T03:53:28Z | cs.CY, cs.CL | This paper presents a groundbreaking comparison between Large Language Models
and traditional legal contract reviewers, Junior Lawyers and Legal Process
Outsourcers. We dissect whether LLMs can outperform humans in accuracy, speed,
and cost efficiency during contract review. Our empirical analysis benchmarks
LLMs against a ground truth set by Senior Lawyers, uncovering that advanced
models match or exceed human accuracy in determining legal issues. In speed,
LLMs complete reviews in mere seconds, eclipsing the hours required by their
human counterparts. Cost wise, LLMs operate at a fraction of the price,
offering a staggering 99.97 percent reduction in cost over traditional methods.
These results are not just statistics, they signal a seismic shift in legal
practice. LLMs stand poised to disrupt the legal industry, enhancing
accessibility and efficiency of legal services. Our research asserts that the
era of LLM dominance in legal contract review is upon us, challenging the
status quo and calling for a reimagined future of legal workflows. | cs.CY | Better Call GPT, Comparing Large Language Models Against Lawyers |
2024-01-24T03:11:36Z | http://arxiv.org/pdf/2402.01695v1 | http://arxiv.org/abs/2402.01695v1 | Alex Zhang, Khanh Nguyen, Jens Tuyls, Albert Lin, Karthik Narasimhan | 2024-01-24T03:11:36Z | cs.CL, cs.AI, cs.LG | Installing probabilistic world models into artificial agents opens an
efficient channel for humans to communicate with and control these agents. In
addition to updating agent policies, humans can modify their internal world
models in order to influence their decisions. The challenge, however, is that
currently existing world models are difficult for humans to adapt because they
lack a natural communication interface. Aimed at addressing this shortcoming,
we develop Language-Guided World Models (LWMs), which can capture environment
dynamics by reading language descriptions. These models enhance agent
communication efficiency, allowing humans to simultaneously alter their
behavior on multiple tasks with concise language feedback. They also enable
agents to self-learn from texts originally written to instruct humans. To
facilitate the development of LWMs, we design a challenging benchmark based on
the game of MESSENGER (Hanjie et al., 2021), requiring compositional
generalization to new language descriptions and environment dynamics. Our
experiments reveal that the current state-of-the-art Transformer architecture
performs poorly on this benchmark, motivating us to design a more robust
architecture. To showcase the practicality of our proposed LWMs, we simulate a
scenario where these models augment the interpretability and safety of an agent
by enabling it to generate and discuss plans with a human before execution. By
effectively incorporating language feedback on the plan, the models boost the
agent performance in the real environment by up to three times without
collecting any interactive experiences in this environment. | cs.CL | Language-Guided World Models: A Model-Based Approach to AI Control |
2024-01-24T03:07:26Z | http://arxiv.org/pdf/2401.13201v3 | http://arxiv.org/abs/2401.13201v3 | Shan Yang, Yongfei Zhang | 2024-06-10T10:21:19Z | cs.CV, cs.AI, cs.CL | Multimodal large language models (MLLM) have achieved satisfactory results in
many tasks. However, their performance in the task of ReID (ReID) has not been
explored to date. This paper will investigate how to adapt them for the task of
ReID. An intuitive idea is to fine-tune MLLM with ReID image-text datasets, and
then use their visual encoder as a backbone for ReID. However, there still
exist two apparent issues: (1) Designing instructions for ReID, MLLMs may
overfit specific instructions, and designing a variety of instructions will
lead to higher costs. (2) When fine-tuning the visual encoder of a MLLM, it is
not trained synchronously with the ReID task. As a result, the effectiveness of
the visual encoder fine-tuning cannot be directly reflected in the performance
of the ReID task. To address these problems, this paper proposes MLLMReID:
Multimodal Large Language Model-based ReID. Firstly, we proposed Common
Instruction, a simple approach that leverages the essence ability of LLMs to
continue writing, avoiding complex and diverse instruction design. Secondly, we
propose a multi-task learning-based synchronization module to ensure that the
visual encoder of the MLLM is trained synchronously with the ReID task. The
experimental results demonstrate the superiority of our method. | cs.CV | MLLMReID: Multimodal Large Language Model-based Person Re-identification |
2024-01-24T01:51:00Z | http://arxiv.org/pdf/2401.13178v1 | http://arxiv.org/abs/2401.13178v1 | Chang Ma, Junlei Zhang, Zhihao Zhu, Cheng Yang, Yujiu Yang, Yaohui Jin, Zhenzhong Lan, Lingpeng Kong, Junxian He | 2024-01-24T01:51:00Z | cs.CL, cs.AI, cs.LG | Evaluating large language models (LLMs) as general-purpose agents is
essential for understanding their capabilities and facilitating their
integration into practical applications. However, the evaluation process
presents substantial challenges. A primary obstacle is the benchmarking of
agent performance across diverse scenarios within a unified framework,
especially in maintaining partially-observable environments and ensuring
multi-round interactions. Moreover, current evaluation frameworks mostly focus
on the final success rate, revealing few insights during the process and
failing to provide a deep understanding of the model abilities. To address
these challenges, we introduce AgentBoard, a pioneering comprehensive benchmark
and accompanied open-source evaluation framework tailored to analytical
evaluation of LLM agents. AgentBoard offers a fine-grained progress rate metric
that captures incremental advancements as well as a comprehensive evaluation
toolkit that features easy assessment of agents for multi-faceted analysis
through interactive visualization. This not only sheds light on the
capabilities and limitations of LLM agents but also propels the
interpretability of their performance to the forefront. Ultimately, AgentBoard
serves as a significant step towards demystifying agent behaviors and
accelerating the development of stronger LLM agents. | cs.CL | AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents |
2024-01-24T01:30:25Z | http://arxiv.org/pdf/2401.13170v4 | http://arxiv.org/abs/2401.13170v4 | Zongxia Li, Ishani Mondal, Yijun Liang, Huy Nghiem, Jordan Boyd-Graber | 2024-06-30T02:09:46Z | cs.CL | Question answering (QA) can only make progress if we know if an answer is
correct, but for many of the most challenging and interesting QA examples,
current evaluation metrics to determine answer equivalence (AE) often do not
align with human judgments, particularly more verbose, free-form answers from
large language models (LLM). There are two challenges: a lack of data and that
models are too big: LLM-based scorers can correlate better with human judges,
but this task has only been tested on limited QA datasets, and even when
available, update of the model is limited because LLMs are large and often
expensive. We rectify both of these issues by providing clear and consistent
guidelines for evaluating AE in machine QA adopted from professional human QA
contests. We also introduce a combination of standard evaluation and a more
efficient, robust, and lightweight discriminate AE classifier-based matching
method (CFMatch, smaller than 1 MB), trained and validated to more accurately
evaluate answer correctness in accordance with adopted expert AE rules that are
more aligned with human judgments. | cs.CL | CFMatch: Aligning Automated Answer Equivalence Evaluation with Expert Judgments For Open-Domain Question Answering |
2024-01-24T00:58:30Z | http://arxiv.org/pdf/2401.13165v3 | http://arxiv.org/abs/2401.13165v3 | Sourojit Ghosh, Srishti Chatterjee | 2024-03-12T17:41:13Z | cs.CL | This chapter focuses on gender-related errors in machine translation (MT) in
the context of low-resource languages. We begin by explaining what low-resource
languages are, examining the inseparable social and computational factors that
create such linguistic hierarchies. We demonstrate through a case study of our
mother tongue Bengali, a global language spoken by almost 300 million people
but still classified as low-resource, how gender is assumed and inferred in
translations to and from the high(est)-resource English when no such
information is provided in source texts. We discuss the postcolonial and
societal impacts of such errors leading to linguistic erasure and
representational harms, and conclude by discussing potential solutions towards
uplifting languages by providing them more agency in MT conversations. | cs.CL | Misgendering and Assuming Gender in Machine Translation when Working with Low-Resource Languages |
2024-01-24T00:36:13Z | http://arxiv.org/pdf/2401.13160v1 | http://arxiv.org/abs/2401.13160v1 | Ke Ye, Heinrich Jiang, Afshin Rostamizadeh, Ayan Chakrabarti, Giulia DeSalvo, Jean-François Kagy, Lazaros Karydas, Gui Citovsky, Sanjiv Kumar | 2024-01-24T00:36:13Z | cs.LG, cs.CL | Pre-training large language models is known to be extremely resource
intensive and often times inefficient, under-utilizing the information
encapsulated in the training text sequences. In this paper, we present SpacTor,
a new training procedure consisting of (1) a hybrid objective combining span
corruption (SC) and token replacement detection (RTD), and (2) a two-stage
curriculum that optimizes the hybrid objective over the initial $\tau$
iterations, then transitions to standard SC loss. We show empirically that the
effectiveness of the hybrid objective is tied to the two-stage pre-training
schedule, and provide extensive analysis on why this is the case. In our
experiments with encoder-decoder architectures (T5) on a variety of NLP tasks,
SpacTor-T5 yields the same downstream performance as standard SC pre-training,
while enabling a 50% reduction in pre-training iterations and 40% reduction in
total FLOPs. Alternatively, given the same amount of computing budget, we find
that SpacTor results in significantly improved downstream benchmark
performance. | cs.LG | SpacTor-T5: Pre-training T5 Models with Span Corruption and Replaced Token Detection |
2024-01-23T23:46:01Z | http://arxiv.org/pdf/2401.13146v1 | http://arxiv.org/abs/2401.13146v1 | Md Asif Jalal, Pablo Peso Parada, George Pavlidis, Vasileios Moschopoulos, Karthikeyan Saravanan, Chrysovalantis-Giorgos Kontoulis, Jisi Zhang, Anastasios Drosou, Gil Ho Lee, Jungin Lee, Seokyeong Jung | 2024-01-23T23:46:01Z | eess.AS, cs.CL, cs.SD | Automatic Speech Recognition (ASR) still face challenges when recognizing
time-variant rare-phrases. Contextual biasing (CB) modules bias ASR model
towards such contextually-relevant phrases. During training, a list of biasing
phrases are selected from a large pool of phrases following a sampling
strategy. In this work we firstly analyse different sampling strategies to
provide insights into the training of CB for ASR with correlation plots between
the bias embeddings among various training stages. Secondly, we introduce a
neighbourhood attention (NA) that localizes self attention (SA) to the nearest
neighbouring frames to further refine the CB output. The results show that this
proposed approach provides on average a 25.84% relative WER improvement on
LibriSpeech sets and rare-word evaluation compared to the baseline. | eess.AS | Locality enhanced dynamic biasing and sampling strategies for contextual ASR |
2024-01-23T23:42:41Z | http://arxiv.org/pdf/2402.01694v1 | http://arxiv.org/abs/2402.01694v1 | Maxim Khanov, Jirayu Burapacheep, Yixuan Li | 2024-01-23T23:42:41Z | cs.CL, cs.AI, cs.LG | Aligning large language models with human objectives is paramount, yet common
approaches including RLHF suffer from unstable and resource-intensive training.
In response to this challenge, we introduce ARGS, Alignment as Reward-Guided
Search, a novel framework that integrates alignment into the decoding process,
eliminating the need for expensive RL training. By adjusting the model's
probabilistic predictions using a reward signal, ARGS generates texts with
semantic diversity while being aligned with human preferences, offering a
promising and flexible solution for aligning language models. Notably, ARGS
demonstrates consistent enhancements in average reward compared to baselines
across diverse alignment tasks and various model dimensions. For example, under
the same greedy-based decoding strategy, our method improves the average reward
by 19.56% relative to the baseline and secures a preference or tie score of
64.33% in GPT-4 evaluation. We believe that our framework, emphasizing
decoding-time alignment, paves the way for more responsive language models in
the future. Code is publicly available at:
\url{https://github.com/deeplearning-wisc/args}. | cs.CL | ARGS: Alignment as Reward-Guided Search |
2024-01-23T23:12:09Z | http://arxiv.org/pdf/2401.13136v1 | http://arxiv.org/abs/2401.13136v1 | Lingfeng Shen, Weiting Tan, Sihao Chen, Yunmo Chen, Jingyu Zhang, Haoran Xu, Boyuan Zheng, Philipp Koehn, Daniel Khashabi | 2024-01-23T23:12:09Z | cs.CL, cs.AI | As the influence of large language models (LLMs) spans across global
communities, their safety challenges in multilingual settings become paramount
for alignment research. This paper examines the variations in safety challenges
faced by LLMs across different languages and discusses approaches to
alleviating such concerns. By comparing how state-of-the-art LLMs respond to
the same set of malicious prompts written in higher- vs. lower-resource
languages, we observe that (1) LLMs tend to generate unsafe responses much more
often when a malicious prompt is written in a lower-resource language, and (2)
LLMs tend to generate more irrelevant responses to malicious prompts in
lower-resource languages. To understand where the discrepancy can be
attributed, we study the effect of instruction tuning with reinforcement
learning from human feedback (RLHF) or supervised finetuning (SFT) on the
HH-RLHF dataset. Surprisingly, while training with high-resource languages
improves model alignment, training in lower-resource languages yields minimal
improvement. This suggests that the bottleneck of cross-lingual alignment is
rooted in the pretraining stage. Our findings highlight the challenges in
cross-lingual LLM safety, and we hope they inform future research in this
direction. | cs.CL | The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts |
2024-01-23T22:49:19Z | http://arxiv.org/pdf/2401.13133v1 | http://arxiv.org/abs/2401.13133v1 | Ibrahim Said Ahmad, Lukman Jibril Aliyu, Abubakar Auwal Khalid, Saminu Muhammad Aliyu, Shamsuddeen Hassan Muhammad, Idris Abdulmumin, Bala Mairiga Abduljalil, Bello Shehu Bello, Amina Imam Abubakar | 2024-01-23T22:49:19Z | cs.CL, cs.SI | Numerous successes have been achieved in combating the COVID-19 pandemic,
initially using various precautionary measures like lockdowns, social
distancing, and the use of face masks. More recently, various vaccinations have
been developed to aid in the prevention or reduction of the severity of the
COVID-19 infection. Despite the effectiveness of the precautionary measures and
the vaccines, there are several controversies that are massively shared on
social media platforms like Twitter. In this paper, we explore the use of
state-of-the-art transformer-based language models to study people's acceptance
of vaccines in Nigeria. We developed a novel dataset by crawling multi-lingual
tweets using relevant hashtags and keywords. Our analysis and visualizations
revealed that most tweets expressed neutral sentiments about COVID-19 vaccines,
with some individuals expressing positive views, and there was no strong
preference for specific vaccine types, although Moderna received slightly more
positive sentiment. We also found out that fine-tuning a pre-trained LLM with
an appropriate dataset can yield competitive results, even if the LLM was not
initially pre-trained on the specific language of that dataset. | cs.CL | Analyzing COVID-19 Vaccination Sentiments in Nigerian Cyberspace: Insights from a Manually Annotated Twitter Dataset |
2024-01-23T22:36:03Z | http://arxiv.org/pdf/2401.13129v2 | http://arxiv.org/abs/2401.13129v2 | Yu Zhang, Yunyi Zhang, Yanzhen Shen, Yu Deng, Lucian Popa, Larisa Shwartz, ChengXiang Zhai, Jiawei Han | 2024-02-20T18:50:46Z | cs.CL, cs.SE | Accurately typing entity mentions from text segments is a fundamental task
for various natural language processing applications. Many previous approaches
rely on massive human-annotated data to perform entity typing. Nevertheless,
collecting such data in highly specialized science and engineering domains
(e.g., software engineering and security) can be time-consuming and costly,
without mentioning the domain gaps between training and inference data if the
model needs to be applied to confidential datasets. In this paper, we study the
task of seed-guided fine-grained entity typing in science and engineering
domains, which takes the name and a few seed entities for each entity type as
the only supervision and aims to classify new entity mentions into both seen
and unseen types (i.e., those without seed entities). To solve this problem, we
propose SEType which first enriches the weak supervision by finding more
entities for each seen type from an unlabeled corpus using the contextualized
representations of pre-trained language models. It then matches the enriched
entities to unlabeled text to get pseudo-labeled samples and trains a textual
entailment model that can make inferences for both seen and unseen types.
Extensive experiments on two datasets covering four domains demonstrate the
effectiveness of SEType in comparison with various baselines. | cs.CL | Seed-Guided Fine-Grained Entity Typing in Science and Engineering Domains |
2024-01-23T22:03:51Z | http://arxiv.org/pdf/2402.01693v1 | http://arxiv.org/abs/2402.01693v1 | Zhe He, Balu Bhasuran, Qiao Jin, Shubo Tian, Karim Hanna, Cindy Shavor, Lisbeth Garcia Arguello, Patrick Murray, Zhiyong Lu | 2024-01-23T22:03:51Z | cs.CL, cs.AI | Lab results are often confusing and hard to understand. Large language models
(LLMs) such as ChatGPT have opened a promising avenue for patients to get their
questions answered. We aim to assess the feasibility of using LLMs to generate
relevant, accurate, helpful, and unharmful responses to lab test-related
questions asked by patients and to identify potential issues that can be
mitigated with augmentation approaches. We first collected lab test results
related question and answer data from Yahoo! Answers and selected 53 QA pairs
for this study. Using the LangChain framework and ChatGPT web portal, we
generated responses to the 53 questions from four LLMs including GPT-4, Meta
LLaMA 2, MedAlpaca, and ORCA_mini. We first assessed the similarity of their
answers using standard QA similarity-based evaluation metrics including ROUGE,
BLEU, METEOR, BERTScore. We also utilized an LLM-based evaluator to judge
whether a target model has higher quality in terms of relevance, correctness,
helpfulness, and safety than the baseline model. Finally, we performed a manual
evaluation with medical experts for all the responses to seven selected
questions on the same four aspects. The results of Win Rate and medical expert
evaluation both showed that GPT-4's responses achieved better scores than all
the other LLM responses and human responses on all four aspects (relevance,
correctness, helpfulness, and safety). However, LLM responses occasionally also
suffer from a lack of interpretation in one's medical context, incorrect
statements, and lack of references. We find that compared to other three LLMs
and human answer from the Q&A website, GPT-4's responses are more accurate,
helpful, relevant, and safer. However, there are cases which GPT-4 responses
are inaccurate and not individualized. We identified a number of ways to
improve the quality of LLM responses. | cs.CL | Quality of Answers of Generative Large Language Models vs Peer Patients for Interpreting Lab Test Results for Lay Patients: Evaluation Study |
2024-01-23T21:55:34Z | http://arxiv.org/pdf/2402.01692v1 | http://arxiv.org/abs/2402.01692v1 | Wei-Ping Huang, Sung-Feng Huang, Hung-yi Lee | 2024-01-23T21:55:34Z | cs.CL, cs.LG | This paper presents an effective transfer learning framework for language
adaptation in text-to-speech systems, with a focus on achieving language
adaptation using minimal labeled and unlabeled data. While many works focus on
reducing the usage of labeled data, very few consider minimizing the usage of
unlabeled data. By utilizing self-supervised features in the pretraining stage,
replacing the noisy portion of pseudo labels with these features during
fine-tuning, and incorporating an embedding initialization trick, our method
leverages more information from unlabeled data compared to conventional
approaches. Experimental results show that our framework is able to synthesize
intelligible speech in unseen languages with only 4 utterances of labeled data
and 15 minutes of unlabeled data. Our methodology continues to surpass
conventional techniques, even when a greater volume of data is accessible.
These findings highlight the potential of our data-efficient language
adaptation framework. | cs.CL | Maximizing Data Efficiency for Cross-Lingual TTS Adaptation by Self-Supervised Representation Mixing and Embedding Initialization |
2024-01-23T20:55:49Z | http://arxiv.org/pdf/2401.13086v1 | http://arxiv.org/abs/2401.13086v1 | Rick Rejeleene, Xiaowei Xu, John Talburt | 2024-01-23T20:55:49Z | cs.CL, cs.AI, cs.LG | Large language models (LLM) are generating information at a rapid pace,
requiring users to increasingly rely and trust the data. Despite remarkable
advances of LLM, Information generated by LLM is not completely trustworthy,
due to challenges in information quality. Specifically, integrity of
Information quality decreases due to unreliable, biased, tokenization during
pre-training of LLM. Moreover, due to decreased information quality issues, has
led towards hallucination, fabricated information. Unreliable information can
lead towards flawed decisions in businesses, which impacts economic activity.
In this work, we introduce novel mathematical information quality evaluation of
LLM, we furthermore analyze and highlight information quality challenges,
scaling laws to systematically scale language models. | cs.CL | Towards Trustable Language Models: Investigating Information Quality of Large Language Models |
2024-01-23T20:54:40Z | http://arxiv.org/pdf/2401.13085v1 | http://arxiv.org/abs/2401.13085v1 | Onkar Litake, Niraj Yagnik, Shreyas Labhsetwar | 2024-01-23T20:54:40Z | cs.CL, cs.AI, cs.LG | Text Augmentation is an important task for low-resource languages. It helps
deal with the problem of data scarcity. A data augmentation strategy is used to
deal with the problem of data scarcity. Through the years, much work has been
done on data augmentation for the English language. In contrast, very less work
has been done on Indian languages. This is contrary to the fact that data
augmentation is used to deal with data scarcity. In this work, we focus on
implementing techniques like Easy Data Augmentation, Back Translation,
Paraphrasing, Text Generation using LLMs, and Text Expansion using LLMs for
text classification on different languages. We focus on 6 Indian languages
namely: Sindhi, Marathi, Hindi, Gujarati, Telugu, and Sanskrit. According to
our knowledge, no such work exists for text augmentation on Indian languages.
We carry out binary as well as multi-class text classification to make our
results more comparable. We get surprising results as basic data augmentation
techniques surpass LLMs. | cs.CL | IndiText Boost: Text Augmentation for Low Resource India Languages |
2024-01-23T19:32:54Z | http://arxiv.org/pdf/2401.13060v1 | http://arxiv.org/abs/2401.13060v1 | Mohammed Alaa Elkomy, Amany Sarhan | 2024-01-23T19:32:54Z | cs.CL, cs.AI | In this paper, we present our approach to tackle Qur'an QA 2023 shared tasks
A and B. To address the challenge of low-resourced training data, we rely on
transfer learning together with a voting ensemble to improve prediction
stability across multiple runs. Additionally, we employ different architectures
and learning mechanisms for a range of Arabic pre-trained transformer-based
models for both tasks. To identify unanswerable questions, we propose using a
thresholding mechanism. Our top-performing systems greatly surpass the baseline
performance on the hidden split, achieving a MAP score of 25.05% for task A and
a partial Average Precision (pAP) of 57.11% for task B. | cs.CL | TCE at Qur'an QA 2023 Shared Task: Low Resource Enhanced Transformer-based Ensemble Approach for Qur'anic QA |
2024-01-23T18:59:43Z | http://arxiv.org/pdf/2401.12975v1 | http://arxiv.org/abs/2401.12975v1 | Qinhong Zhou, Sunli Chen, Yisong Wang, Haozhe Xu, Weihua Du, Hongxin Zhang, Yilun Du, Joshua B. Tenenbaum, Chuang Gan | 2024-01-23T18:59:43Z | cs.CV, cs.AI, cs.CL | Recent advances in high-fidelity virtual environments serve as one of the
major driving forces for building intelligent embodied agents to perceive,
reason and interact with the physical world. Typically, these environments
remain unchanged unless agents interact with them. However, in real-world
scenarios, agents might also face dynamically changing environments
characterized by unexpected events and need to rapidly take action accordingly.
To remedy this gap, we propose a new simulated embodied benchmark, called
HAZARD, specifically designed to assess the decision-making abilities of
embodied agents in dynamic situations. HAZARD consists of three unexpected
disaster scenarios, including fire, flood, and wind, and specifically supports
the utilization of large language models (LLMs) to assist common sense
reasoning and decision-making. This benchmark enables us to evaluate autonomous
agents' decision-making capabilities across various pipelines, including
reinforcement learning (RL), rule-based, and search-based methods in
dynamically changing environments. As a first step toward addressing this
challenge using large language models, we further develop an LLM-based agent
and perform an in-depth analysis of its promise and challenge of solving these
challenging tasks. HAZARD is available at https://vis-www.cs.umass.edu/hazard/. | cs.CV | HAZARD Challenge: Embodied Decision Making in Dynamically Changing Environments |
2024-01-23T18:59:21Z | http://arxiv.org/pdf/2401.12973v2 | http://arxiv.org/abs/2401.12973v2 | Ekin Akyürek, Bailin Wang, Yoon Kim, Jacob Andreas | 2024-01-30T18:59:34Z | cs.CL, cs.LG | Large-scale neural language models exhibit a remarkable capacity for
in-context learning (ICL): they can infer novel functions from datasets
provided as input. Most of our current understanding of when and how ICL arises
comes from LMs trained on extremely simple learning problems like linear
regression and associative recall. There remains a significant gap between
these model problems and the "real" ICL exhibited by LMs trained on large text
corpora, which involves not just retrieval and function approximation but
free-form generation of language and other structured outputs. In this paper,
we study ICL through the lens of a new family of model problems we term in
context language learning (ICLL). In ICLL, LMs are presented with a set of
strings from a formal language, and must generate additional strings from the
same language. We focus on in-context learning of regular languages generated
by random finite automata. We evaluate a diverse set of neural sequence models
(including several RNNs, Transformers, and state-space model variants) on
regular ICLL tasks, aiming to answer three questions: (1) Which model classes
are empirically capable of ICLL? (2) What algorithmic solutions do successful
models implement to perform ICLL? (3) What architectural changes can improve
ICLL in less performant models? We first show that Transformers significantly
outperform neural sequence models with recurrent or convolutional
representations on ICLL tasks. Next, we provide evidence that their ability to
do so relies on specialized "n-gram heads" (higher-order variants of induction
heads) that compute input-conditional next-token distributions. Finally, we
show that hard-wiring these heads into neural models improves performance not
just on ICLL, but natural language modeling -- improving the perplexity of
340M-parameter models by up to 1.14 points (6.7%) on the SlimPajama dataset. | cs.CL | In-Context Language Learning: Architectures and Algorithms |
2024-01-23T18:57:53Z | http://arxiv.org/pdf/2401.12970v2 | http://arxiv.org/abs/2401.12970v2 | Chengzhi Mao, Carl Vondrick, Hao Wang, Junfeng Yang | 2024-04-14T22:34:37Z | cs.CL | We find that large language models (LLMs) are more likely to modify
human-written text than AI-generated text when tasked with rewriting. This
tendency arises because LLMs often perceive AI-generated text as high-quality,
leading to fewer modifications. We introduce a method to detect AI-generated
content by prompting LLMs to rewrite text and calculating the editing distance
of the output. We dubbed our geneRative AI Detection viA Rewriting method
Raidar. Raidar significantly improves the F1 detection scores of existing AI
content detection models -- both academic and commercial -- across various
domains, including News, creative writing, student essays, code, Yelp reviews,
and arXiv papers, with gains of up to 29 points. Operating solely on word
symbols without high-dimensional features, our method is compatible with black
box LLMs, and is inherently robust on new content. Our results illustrate the
unique imprint of machine-generated text through the lens of the machines
themselves. | cs.CL | Raidar: geneRative AI Detection viA Rewriting |
2024-01-23T18:45:54Z | http://arxiv.org/pdf/2401.12963v2 | http://arxiv.org/abs/2401.12963v2 | Michael Ahn, Debidatta Dwibedi, Chelsea Finn, Montse Gonzalez Arenas, Keerthana Gopalakrishnan, Karol Hausman, Brian Ichter, Alex Irpan, Nikhil Joshi, Ryan Julian, Sean Kirmani, Isabel Leal, Edward Lee, Sergey Levine, Yao Lu, Isabel Leal, Sharath Maddineni, Kanishka Rao, Dorsa Sadigh, Pannag Sanketi, Pierre Sermanet, Quan Vuong, Stefan Welker, Fei Xia, Ted Xiao, Peng Xu, Steve Xu, Zhuo Xu | 2024-07-02T01:52:26Z | cs.RO, cs.AI, cs.CL, cs.CV, cs.LG | Foundation models that incorporate language, vision, and more recently
actions have revolutionized the ability to harness internet scale data to
reason about useful tasks. However, one of the key challenges of training
embodied foundation models is the lack of data grounded in the physical world.
In this paper, we propose AutoRT, a system that leverages existing foundation
models to scale up the deployment of operational robots in completely unseen
scenarios with minimal human supervision. AutoRT leverages vision-language
models (VLMs) for scene understanding and grounding, and further uses large
language models (LLMs) for proposing diverse and novel instructions to be
performed by a fleet of robots. Guiding data collection by tapping into the
knowledge of foundation models enables AutoRT to effectively reason about
autonomy tradeoffs and safety while significantly scaling up data collection
for robot learning. We demonstrate AutoRT proposing instructions to over 20
robots across multiple buildings and collecting 77k real robot episodes via
both teleoperation and autonomous robot policies. We experimentally show that
such "in-the-wild" data collected by AutoRT is significantly more diverse, and
that AutoRT's use of LLMs allows for instruction following data collection
robots that can align to human preferences. | cs.RO | AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic Agents |
2024-01-23T18:22:19Z | http://arxiv.org/pdf/2401.12954v1 | http://arxiv.org/abs/2401.12954v1 | Mirac Suzgun, Adam Tauman Kalai | 2024-01-23T18:22:19Z | cs.CL, cs.AI, cs.HC | We introduce meta-prompting, an effective scaffolding technique designed to
enhance the functionality of language models (LMs). This approach transforms a
single LM into a multi-faceted conductor, adept at managing and integrating
multiple independent LM queries. By employing high-level instructions,
meta-prompting guides the LM to break down complex tasks into smaller, more
manageable subtasks. These subtasks are then handled by distinct "expert"
instances of the same LM, each operating under specific, tailored instructions.
Central to this process is the LM itself, in its role as the conductor, which
ensures seamless communication and effective integration of the outputs from
these expert models. It additionally employs its inherent critical thinking and
robust verification processes to refine and authenticate the end result. This
collaborative prompting approach empowers a single LM to simultaneously act as
a comprehensive orchestrator and a panel of diverse experts, significantly
enhancing its performance across a wide array of tasks. The zero-shot,
task-agnostic nature of meta-prompting greatly simplifies user interaction by
obviating the need for detailed, task-specific instructions. Furthermore, our
research demonstrates the seamless integration of external tools, such as a
Python interpreter, into the meta-prompting framework, thereby broadening its
applicability and utility. Through rigorous experimentation with GPT-4, we
establish the superiority of meta-prompting over conventional scaffolding
methods: When averaged across all tasks, including the Game of 24,
Checkmate-in-One, and Python Programming Puzzles, meta-prompting, augmented
with a Python interpreter functionality, surpasses standard prompting by 17.1%,
expert (dynamic) prompting by 17.3%, and multipersona prompting by 15.2%. | cs.CL | Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffolding |
2024-01-23T18:07:38Z | http://arxiv.org/pdf/2401.12947v1 | http://arxiv.org/abs/2401.12947v1 | Dylan Zhang, Curt Tigges, Zory Zhang, Stella Biderman, Maxim Raginsky, Talia Ringer | 2024-01-23T18:07:38Z | cs.CL, cs.AI, cs.FL, cs.LO, cs.PL | This paper investigates the ability of transformer-based models to learn
structural recursion from examples. Recursion is a universal concept in both
natural and formal languages. Structural recursion is central to the
programming language and formal mathematics tasks where symbolic tools
currently excel beyond neural models, such as inferring semantic relations
between datatypes and emulating program behavior. We introduce a general
framework that nicely connects the abstract concepts of structural recursion in
the programming language domain to concrete sequence modeling problems and
learned models' behavior. The framework includes a representation that captures
the general \textit{syntax} of structural recursion, coupled with two different
frameworks for understanding their \textit{semantics} -- one that is more
natural from a programming languages perspective and one that helps bridge that
perspective with a mechanistic understanding of the underlying transformer
architecture.
With our framework as a powerful conceptual tool, we identify different
issues under various set-ups. The models trained to emulate recursive
computations cannot fully capture the recursion yet instead fit short-cut
algorithms and thus cannot solve certain edge cases that are under-represented
in the training distribution. In addition, it is difficult for state-of-the-art
large language models (LLMs) to mine recursive rules from in-context
demonstrations. Meanwhile, these LLMs fail in interesting ways when emulating
reduction (step-wise computation) of the recursive function. | cs.CL | Transformer-Based Models Are Not Yet Perfect At Learning to Emulate Structural Recursion |
2024-01-23T17:58:38Z | http://arxiv.org/pdf/2401.12941v1 | http://arxiv.org/abs/2401.12941v1 | Alexandra Loessberg-Zahl | 2024-01-23T17:58:38Z | cs.CL | State of the art Named Entity Recognition (NER) models have achieved an
impressive ability to extract common phrases from text that belong to labels
such as location, organization, time, and person. However, typical NER systems
that rely on having seen a specific entity in their training data in order to
label an entity perform poorly on rare or unseen entities ta in order to label
an entity perform poorly on rare or unseen entities (Derczynski et al., 2017).
This paper attempts to improve recognition of person names, a diverse category
that can grow any time someone is born or changes their name. In order for
downstream tasks to not exhibit bias based on cultural background, a model
should perform well on names from a variety of backgrounds. In this paper I
experiment with the training data and input structure of an English Bi-LSTM
name recognition model. I look at names from 103 countries to compare how well
the model performs on names from different cultures, specifically in the
context of a downstream task where extracted names will be matched to
information on file. I find that a model with combined character and word input
outperforms word-only models and may improve on accuracy compared to classical
NER models that are not geared toward identifying unseen entity values. | cs.CL | Multicultural Name Recognition For Previously Unseen Names |
2024-01-23T17:07:18Z | http://arxiv.org/pdf/2401.12915v1 | http://arxiv.org/abs/2401.12915v1 | Mukai Li, Lei Li, Yuwei Yin, Masood Ahmed, Zhenguang Liu, Qi Liu | 2024-01-23T17:07:18Z | cs.AI, cs.CL, cs.CV | VLMs (Vision-Language Models) extend the capabilities of LLMs (Large Language
Models) to accept multimodal inputs. Since it has been verified that LLMs can
be induced to generate harmful or inaccurate content through specific test
cases (termed as Red Teaming), how VLMs perform in similar scenarios,
especially with their combination of textual and visual inputs, remains a
question. To explore this problem, we present a novel red teaming dataset
RTVLM, which encompasses 10 subtasks (e.g., image misleading, multi-modal
jail-breaking, face fairness, etc) under 4 primary aspects (faithfulness,
privacy, safety, fairness). Our RTVLM is the first red-teaming dataset to
benchmark current VLMs in terms of these 4 different aspects. Detailed analysis
shows that 10 prominent open-sourced VLMs struggle with the red teaming in
different degrees and have up to 31% performance gap with GPT-4V. Additionally,
we simply apply red teaming alignment to LLaVA-v1.5 with Supervised Fine-tuning
(SFT) using RTVLM, and this bolsters the models' performance with 10% in RTVLM
test set, 13% in MM-Hal, and without noticeable decline in MM-Bench,
overpassing other LLaVA-based models with regular alignment data. This reveals
that current open-sourced VLMs still lack red teaming alignment. Our code and
datasets will be open-source. | cs.AI | Red Teaming Visual Language Models |
2024-01-23T16:30:22Z | http://arxiv.org/pdf/2402.01690v1 | http://arxiv.org/abs/2402.01690v1 | Ali Pourramezan Fard, Mohammad H. Mahoor, Muath Alsuhaibani, Hiroko H. Dodgec | 2024-01-23T16:30:22Z | cs.CL, cs.LG | This paper presents a deep learning method using Natural Language Processing
(NLP) techniques, to distinguish between Mild Cognitive Impairment (MCI) and
Normal Cognitive (NC) conditions in older adults. We propose a framework that
analyzes transcripts generated from video interviews collected within the
I-CONECT study project, a randomized controlled trial aimed at improving
cognitive functions through video chats. Our proposed NLP framework consists of
two Transformer-based modules, namely Sentence Embedding (SE) and Sentence
Cross Attention (SCA). First, the SE module captures contextual relationships
between words within each sentence. Subsequently, the SCA module extracts
temporal features from a sequence of sentences. This feature is then used by a
Multi-Layer Perceptron (MLP) for the classification of subjects into MCI or NC.
To build a robust model, we propose a novel loss function, called InfoLoss,
that considers the reduction in entropy by observing each sequence of sentences
to ultimately enhance the classification accuracy. The results of our
comprehensive model evaluation using the I-CONECT dataset show that our
framework can distinguish between MCI and NC with an average area under the
curve of 84.75%. | cs.CL | Linguistic-Based Mild Cognitive Impairment Detection Using Informative Loss |
2024-01-23T16:09:53Z | http://arxiv.org/pdf/2401.12874v2 | http://arxiv.org/abs/2401.12874v2 | Haoyan Luo, Lucia Specia | 2024-02-22T04:28:03Z | cs.CL, cs.AI | Explainability for Large Language Models (LLMs) is a critical yet challenging
aspect of natural language processing. As LLMs are increasingly integral to
diverse applications, their "black-box" nature sparks significant concerns
regarding transparency and ethical use. This survey underscores the imperative
for increased explainability in LLMs, delving into both the research on
explainability and the various methodologies and tasks that utilize an
understanding of these models. Our focus is primarily on pre-trained
Transformer-based LLMs, such as LLaMA family, which pose distinctive
interpretability challenges due to their scale and complexity. In terms of
existing methods, we classify them into local and global analyses, based on
their explanatory objectives. When considering the utilization of
explainability, we explore several compelling methods that concentrate on model
editing, control generation, and model enhancement. Additionally, we examine
representative evaluation metrics and datasets, elucidating their advantages
and limitations. Our goal is to reconcile theoretical and empirical
understanding with practical implementation, proposing exciting avenues for
explanatory techniques and their applications in the LLMs era. | cs.CL | From Understanding to Utilization: A Survey on Explainability for Large Language Models |
2024-01-23T16:07:43Z | http://arxiv.org/pdf/2401.12873v3 | http://arxiv.org/abs/2401.12873v3 | Zhiwei He, Xing Wang, Wenxiang Jiao, Zhuosheng Zhang, Rui Wang, Shuming Shi, Zhaopeng Tu | 2024-03-18T15:16:16Z | cs.CL, cs.AI | Insufficient modeling of human preferences within the reward model is a major
obstacle for leveraging human feedback to improve translation quality.
Fortunately, quality estimation (QE), which predicts the quality of a given
translation without reference, has achieved impressive alignment with human
evaluations in the last two years. In this work, we investigate the potential
of employing the QE model as the reward model to predict human preferences for
feedback training. We first identify the overoptimization problem during
QE-based feedback training, manifested as an increase in reward while
translation quality declines. We examine the problem and argue that the
vulnerability of the QE model might lead to high rewards for incorrect
translations, resulting in overoptimization and error propagation. To address
the problem, we adopt a simple yet effective method that uses heuristic rules
to detect the incorrect translations and assigns a penalty term to the reward
scores of them. Experimental results show that the proposed QE-based feedback
training achieves consistent and significant improvements across various
settings, further verified through human preference studies. Our subsequent
analysis demonstrates the high data efficiency of the proposed QE-based
feedback training: it outperforms systems using larger parallel corpora by a
small amount of monolingual data. Our code is available at:
https://github.com/zwhe99/FeedbackMT | cs.CL | Improving Machine Translation with Human Feedback: An Exploration of Quality Estimation as a Reward Model |
2024-01-23T15:56:11Z | http://arxiv.org/pdf/2401.12863v1 | http://arxiv.org/abs/2401.12863v1 | Debjyoti Mondal, Suraj Modi, Subhadarshi Panda, Rituraj Singh, Godawari Sudhakar Rao | 2024-01-23T15:56:11Z | cs.CL, cs.AI | Large Language Models (LLMs) have demonstrated impressive performance in
natural language processing tasks by leveraging chain of thought (CoT) that
enables step-by-step thinking. Extending LLMs with multimodal capabilities is
the recent interest, but incurs computational cost and requires substantial
hardware resources. To address these challenges, we propose KAM-CoT a framework
that integrates CoT reasoning, Knowledge Graphs (KGs), and multiple modalities
for a comprehensive understanding of multimodal tasks. KAM-CoT adopts a
two-stage training process with KG grounding to generate effective rationales
and answers. By incorporating external knowledge from KGs during reasoning, the
model gains a deeper contextual understanding reducing hallucinations and
enhancing the quality of answers. This knowledge-augmented CoT reasoning
empowers the model to handle questions requiring external context, providing
more informed answers. Experimental findings show KAM-CoT outperforms the
state-of-the-art methods. On the ScienceQA dataset, we achieve an average
accuracy of 93.87%, surpassing GPT-3.5 (75.17%) by 18% and GPT-4 (83.99%) by
10%. Remarkably, KAM-CoT achieves these results with only 280M trainable
parameters at a time, demonstrating its cost-efficiency and effectiveness. | cs.CL | KAM-CoT: Knowledge Augmented Multimodal Chain-of-Thoughts Reasoning |
2024-01-23T14:31:12Z | http://arxiv.org/pdf/2401.12798v3 | http://arxiv.org/abs/2401.12798v3 | Yuanyi Wang, Haifeng Sun, Jingyu Wang, Qi Qi, Shaoling Sun, Jianxin Liao | 2024-04-17T17:00:34Z | cs.IR, cs.CL | Entity alignment (EA), a pivotal process in integrating multi-source
Knowledge Graphs (KGs), seeks to identify equivalent entity pairs across these
graphs. Most existing approaches regard EA as a graph representation learning
task, concentrating on enhancing graph encoders. However, the decoding process
in EA - essential for effective operation and alignment accuracy - has received
limited attention and remains tailored to specific datasets and model
architectures, necessitating both entity and additional explicit relation
embeddings. This specificity limits its applicability, particularly in
GNN-based models. To address this gap, we introduce a novel, generalized, and
efficient decoding approach for EA, relying solely on entity embeddings. Our
method optimizes the decoding process by minimizing Dirichlet energy, leading
to the gradient flow within the graph, to maximize graph homophily. The
discretization of the gradient flow produces a fast and scalable approach,
termed Triple Feature Propagation (TFP). TFP innovatively generalizes adjacency
matrices to multi-views matrices:entity-to-entity, entity-to-relation,
relation-to-entity, and relation-to-triple. The gradient flow through
generalized matrices enables TFP to harness the multi-view structural
information of KGs. Rigorous experimentation on diverse public datasets
demonstrates that our approach significantly enhances various EA methods.
Notably, the approach achieves these advancements with less than 6 seconds of
additional computational time, establishing a new benchmark in efficiency and
adaptability for future EA methods. | cs.IR | Gradient Flow of Energy: A General and Efficient Approach for Entity Alignment Decoding |
2024-01-23T14:29:17Z | http://arxiv.org/pdf/2401.12794v2 | http://arxiv.org/abs/2401.12794v2 | Fanghua Ye, Mingming Yang, Jianhui Pang, Longyue Wang, Derek F. Wong, Emine Yilmaz, Shuming Shi, Zhaopeng Tu | 2024-04-25T14:00:01Z | cs.CL | The proliferation of open-source Large Language Models (LLMs) from various
institutions has highlighted the urgent need for comprehensive evaluation
methods. However, current evaluation platforms, such as the widely recognized
HuggingFace open LLM leaderboard, neglect a crucial aspect -- uncertainty,
which is vital for thoroughly assessing LLMs. To bridge this gap, we introduce
a new benchmarking approach for LLMs that integrates uncertainty
quantification. Our examination involves eight LLMs (LLM series) spanning five
representative natural language processing tasks. Our findings reveal that: I)
LLMs with higher accuracy may exhibit lower certainty; II) Larger-scale LLMs
may display greater uncertainty compared to their smaller counterparts; and
III) Instruction-finetuning tends to increase the uncertainty of LLMs. These
results underscore the significance of incorporating uncertainty in the
evaluation of LLMs. | cs.CL | Benchmarking LLMs via Uncertainty Quantification |
2024-01-23T14:19:01Z | http://arxiv.org/pdf/2401.12789v1 | http://arxiv.org/abs/2401.12789v1 | W. Ronny Huang, Cyril Allauzen, Tongzhou Chen, Kilol Gupta, Ke Hu, James Qin, Yu Zhang, Yongqiang Wang, Shuo-Yiin Chang, Tara N. Sainath | 2024-01-23T14:19:01Z | cs.CL, cs.SD, eess.AS | In the era of large models, the autoregressive nature of decoding often
results in latency serving as a significant bottleneck. We propose a
non-autoregressive LM-fused ASR system that effectively leverages the
parallelization capabilities of accelerator hardware. Our approach combines the
Universal Speech Model (USM) and the PaLM 2 language model in per-segment
scoring mode, achieving an average relative WER improvement across all
languages of 10.8% on FLEURS and 3.6% on YouTube captioning. Furthermore, our
comprehensive ablation study analyzes key parameters such as LLM size, context
length, vocabulary size, fusion methodology. For instance, we explore the
impact of LLM size ranging from 128M to 340B parameters on ASR performance.
This study provides valuable insights into the factors influencing the
effectiveness of practical large-scale LM-fused speech recognition systems. | cs.CL | Multilingual and Fully Non-Autoregressive ASR with Large Language Model Fusion: A Comprehensive Study |
2024-01-23T13:35:47Z | http://arxiv.org/pdf/2401.12756v2 | http://arxiv.org/abs/2401.12756v2 | Carolin Holtermann, Markus Frohmann, Navid Rekabsaz, Anne Lauscher | 2024-01-25T14:32:36Z | cs.CL, cs.AI | The knowledge encapsulated in a model is the core factor determining its
final performance on downstream tasks. Much research in NLP has focused on
efficient methods for storing and adapting different types of knowledge, e.g.,
in dedicated modularized structures, and on how to effectively combine these,
e.g., by learning additional parameters. However, given the many possible
options, a thorough understanding of the mechanisms involved in these
compositions is missing, and hence it remains unclear which strategies to
utilize. To address this research gap, we propose a novel framework for
zero-shot module composition, which encompasses existing and some novel
variations for selecting, weighting, and combining parameter modules under a
single unified notion. Focusing on the scenario of domain knowledge and adapter
layers, our framework provides a systematic unification of concepts, allowing
us to conduct the first comprehensive benchmarking study of various zero-shot
knowledge composition strategies. In particular, we test two module combination
methods and five selection and weighting strategies for their effectiveness and
efficiency in an extensive experimental setup. Our results highlight the
efficacy of ensembling but also hint at the power of simple though
often-ignored weighting methods. Further in-depth analyses allow us to
understand the role of weighting vs. top-k selection, and show that, to a
certain extent, the performance of adapter composition can even be predicted. | cs.CL | What the Weight?! A Unified Framework for Zero-Shot Knowledge Composition |
2024-01-23T12:41:03Z | http://arxiv.org/pdf/2401.12720v1 | http://arxiv.org/abs/2401.12720v1 | Guilherme H. Resende, Luiz F. Nery, Fabrício Benevenuto, Savvas Zannettou, Flavio Figueiredo | 2024-01-23T12:41:03Z | cs.CL, cs.SI | Language is a dynamic aspect of our culture that changes when expressed in
different technologies/communities. Online social networks have enabled the
diffusion and evolution of different dialects, including African American
English (AAE). However, this increased usage is not without barriers. One
particular barrier is how sentiment (Vader, TextBlob, and Flair) and toxicity
(Google's Perspective and the open-source Detoxify) methods present biases
towards utterances with AAE expressions. Consider Google's Perspective to
understand bias. Here, an utterance such as ``All n*ggers deserve to die
respectfully. The police murder us.'' it reaches a higher toxicity than
``African-Americans deserve to die respectfully. The police murder us.''. This
score difference likely arises because the tool cannot understand the
re-appropriation of the term ``n*gger''. One explanation for this bias is that
AI models are trained on limited datasets, and using such a term in training
data is more likely to appear in a toxic utterance. While this may be
plausible, the tool will make mistakes regardless. Here, we study bias on two
Web-based (YouTube and Twitter) datasets and two spoken English datasets. Our
analysis shows how most models present biases towards AAE in most settings. We
isolate the impact of AAE expression usage via linguistic control features from
the Linguistic Inquiry and Word Count (LIWC) software, grammatical control
features extracted via Part-of-Speech (PoS) tagging from Natural Language
Processing (NLP) models, and the semantic of utterances by comparing sentence
embeddings from recent language models. We present consistent results on how a
heavy usage of AAE expressions may cause the speaker to be considered
substantially more toxic, even when speaking about nearly the same subject. Our
study complements similar analyses focusing on small datasets and/or one method
only. | cs.CL | A Comprehensive View of the Biases of Toxicity and Sentiment Analysis Methods Towards Utterances with African American English Expressions |
2024-01-23T12:29:37Z | http://arxiv.org/pdf/2401.12713v3 | http://arxiv.org/abs/2401.12713v3 | Iman Munire Bilal, Preslav Nakov, Rob Procter, Maria Liakata | 2024-02-23T15:01:38Z | cs.CL | The task of rumour verification in social media concerns assessing the
veracity of a claim on the basis of conversation threads that result from it.
While previous work has focused on predicting a veracity label, here we
reformulate the task to generate model-centric free-text explanations of a
rumour's veracity. The approach is model agnostic in that it generalises to any
model. Here we propose a novel GNN-based rumour verification model. We follow a
zero-shot approach by first applying post-hoc explainability methods to score
the most important posts within a thread and then we use these posts to
generate informative explanations using opinion-guided summarisation. To
evaluate the informativeness of the explanatory summaries, we exploit the
few-shot learning capabilities of a large language model (LLM). Our experiments
show that LLMs can have similar agreement to humans in evaluating summaries.
Importantly, we show explanatory abstractive summaries are more informative and
better reflect the predicted rumour veracity than just using the highest
ranking posts in the thread. | cs.CL | Generating Zero-shot Abstractive Explanations for Rumour Verification |
2024-01-23T11:54:09Z | http://arxiv.org/pdf/2401.12689v3 | http://arxiv.org/abs/2401.12689v3 | Ru Peng, Heming Zou, Haobo Wang, Yawen Zeng, Zenan Huang, Junbo Zhao | 2024-03-15T06:51:28Z | cs.LG, cs.AI, cs.CL, cs.CV | The conventional evaluation protocols on machine learning models rely heavily
on a labeled, i.i.d-assumed testing dataset, which is not often present in real
world applications. The Automated Model Evaluation (AutoEval) shows an
alternative to this traditional workflow, by forming a proximal prediction
pipeline of the testing performance without the presence of ground-truth
labels. Despite its recent successes, the AutoEval frameworks still suffer from
an overconfidence issue, substantial storage and computational cost. In that
regard, we propose a novel measure -- Meta-Distribution Energy (MDE) -- that
allows the AutoEval framework to be both more efficient and effective. The core
of the MDE is to establish a meta-distribution statistic, on the information
(energy) associated with individual samples, then offer a smoother
representation enabled by energy-based learning. We further provide our
theoretical insights by connecting the MDE with the classification loss. We
provide extensive experiments across modalities, datasets and different
architectural backbones to validate MDE's validity, together with its
superiority compared with prior approaches. We also prove MDE's versatility by
showing its seamless integration with large-scale models, and easy adaption to
learning scenarios with noisy- or imbalanced- labels. Code and data are
available: https://github.com/pengr/Energy_AutoEval | cs.LG | Energy-based Automated Model Evaluation |
2024-01-23T11:25:34Z | http://arxiv.org/pdf/2401.12671v2 | http://arxiv.org/abs/2401.12671v2 | Somnath Banerjee, Amruit Sahoo, Sayan Layek, Avik Dutta, Rima Hazra, Animesh Mukherjee | 2024-03-05T07:18:53Z | cs.CL | In the continuously advancing AI landscape, crafting context-rich and
meaningful responses via Large Language Models (LLMs) is essential. Researchers
are becoming more aware of the challenges that LLMs with fewer parameters
encounter when trying to provide suitable answers to open-ended questions. To
address these hurdles, the integration of cutting-edge strategies, augmentation
of rich external domain knowledge to LLMs, offers significant improvements.
This paper introduces a novel framework that combines graph-driven context
retrieval in conjunction to knowledge graphs based enhancement, honing the
proficiency of LLMs, especially in domain specific community question answering
platforms like AskUbuntu, Unix, and ServerFault. We conduct experiments on
various LLMs with different parameter sizes to evaluate their ability to ground
knowledge and determine factual accuracy in answers to open-ended questions.
Our methodology GraphContextGen consistently outperforms dominant text-based
retrieval systems, demonstrating its robustness and adaptability to a larger
number of use cases. This advancement highlights the importance of pairing
context rich data retrieval with LLMs, offering a renewed approach to knowledge
sourcing and generation in AI systems. We also show that, due to rich
contextual data retrieval, the crucial entities, along with the generated
answer, remain factually coherent with the gold answer. | cs.CL | Context Matters: Pushing the Boundaries of Open-Ended Answer Generation with Graph-Structured Knowledge Context |
2024-01-23T10:27:42Z | http://arxiv.org/pdf/2401.12631v1 | http://arxiv.org/abs/2401.12631v1 | Zhengxuan Wu, Atticus Geiger, Jing Huang, Aryaman Arora, Thomas Icard, Christopher Potts, Noah D. Goodman | 2024-01-23T10:27:42Z | cs.LG, cs.AI, cs.CL | We respond to the recent paper by Makelov et al. (2023), which reviews
subspace interchange intervention methods like distributed alignment search
(DAS; Geiger et al. 2023) and claims that these methods potentially cause
"interpretability illusions". We first review Makelov et al. (2023)'s technical
notion of what an "interpretability illusion" is, and then we show that even
intuitive and desirable explanations can qualify as illusions in this sense. As
a result, their method of discovering "illusions" can reject explanations they
consider "non-illusory". We then argue that the illusions Makelov et al. (2023)
see in practice are artifacts of their training and evaluation paradigms. We
close by emphasizing that, though we disagree with their core characterization,
Makelov et al. (2023)'s examples and discussion have undoubtedly pushed the
field of interpretability forward. | cs.LG | A Reply to Makelov et al. (2023)'s "Interpretability Illusion" Arguments |
2024-01-23T09:33:31Z | http://arxiv.org/pdf/2401.12585v4 | http://arxiv.org/abs/2401.12585v4 | Lingrui Mei, Shenghua Liu, Yiwei Wang, Baolong Bi, Xueqi Cheng | 2024-02-20T08:54:50Z | cs.CL | The dynamic nature of language, particularly evident in the realm of slang
and memes on the Internet, poses serious challenges to the adaptability of
large language models (LLMs). Traditionally anchored to static datasets, these
models often struggle to keep up with the rapid linguistic evolution
characteristic of online communities. This research aims to bridge this gap by
enhancing LLMs' comprehension of the evolving new concepts on the Internet,
without the high cost of continual retraining. In pursuit of this goal, we
introduce $\textbf{SLANG}$, a benchmark designed to autonomously integrate
novel data and assess LLMs' ability to comprehend emerging concepts, alongside
$\textbf{FOCUS}$, an approach uses causal inference to enhance LLMs to
understand new phrases and their colloquial context. Our benchmark and approach
involves understanding real-world instances of linguistic shifts, serving as
contextual beacons, to form more precise and contextually relevant connections
between newly emerging expressions and their meanings. The empirical analysis
shows that our causal inference-based approach outperforms the baseline methods
in terms of precision and relevance in the comprehension of Internet slang and
memes. | cs.CL | SLANG: New Concept Comprehension of Large Language Models |
2024-01-23T09:11:07Z | http://arxiv.org/pdf/2401.12576v2 | http://arxiv.org/abs/2401.12576v2 | Qianli Wang, Tatiana Anikina, Nils Feldhus, Josef van Genabith, Leonhard Hennig, Sebastian Möller | 2024-04-24T17:17:48Z | cs.CL, cs.AI, cs.LG | Interpretability tools that offer explanations in the form of a dialogue have
demonstrated their efficacy in enhancing users' understanding (Slack et al.,
2023; Shen et al., 2023), as one-off explanations may fall short in providing
sufficient information to the user. Current solutions for dialogue-based
explanations, however, often require external tools and modules and are not
easily transferable to tasks they were not designed for. With LLMCheckup, we
present an easily accessible tool that allows users to chat with any
state-of-the-art large language model (LLM) about its behavior. We enable LLMs
to generate explanations and perform user intent recognition without
fine-tuning, by connecting them with a broad spectrum of Explainable AI (XAI)
methods, including white-box explainability tools such as feature attributions,
and self-explanations (e.g., for rationale generation). LLM-based
(self-)explanations are presented as an interactive dialogue that supports
follow-up questions and generates suggestions. LLMCheckupprovides tutorials for
operations available in the system, catering to individuals with varying levels
of expertise in XAI and supporting multiple input modalities. We introduce a
new parsing strategy that substantially enhances the user intent recognition
accuracy of the LLM. Finally, we showcase LLMCheckup for the tasks of fact
checking and commonsense question answering. | cs.CL | LLMCheckup: Conversational Examination of Large Language Models via Interpretability Tools and Self-Explanations |
2024-01-23T08:49:23Z | http://arxiv.org/pdf/2401.12566v1 | http://arxiv.org/abs/2401.12566v1 | Markus Leippold, Saeid Ashraf Vaghefi, Dominik Stammbach, Veruska Muccione, Julia Bingler, Jingwei Ni, Chiara Colesanti-Senni, Tobias Wekhof, Tobias Schimanski, Glen Gostlow, Tingyu Yu, Juerg Luterbacher, Christian Huggel | 2024-01-23T08:49:23Z | cs.CL | This paper presents Climinator, a novel AI-based tool designed to automate
the fact-checking of climate change claims. Utilizing an array of Large
Language Models (LLMs) informed by authoritative sources like the IPCC reports
and peer-reviewed scientific literature, Climinator employs an innovative
Mediator-Advocate framework. This design allows Climinator to effectively
synthesize varying scientific perspectives, leading to robust, evidence-based
evaluations. Our model demonstrates remarkable accuracy when testing claims
collected from Climate Feedback and Skeptical Science. Notably, when
integrating an advocate with a climate science denial perspective in our
framework, Climinator's iterative debate process reliably converges towards
scientific consensus, underscoring its adeptness at reconciling diverse
viewpoints into science-based, factual conclusions. While our research is
subject to certain limitations and necessitates careful interpretation, our
approach holds significant potential. We hope to stimulate further research and
encourage exploring its applicability in other contexts, including political
fact-checking and legal domains. | cs.CL | Automated Fact-Checking of Climate Change Claims with Large Language Models |
2024-01-23T07:48:58Z | http://arxiv.org/pdf/2401.12540v1 | http://arxiv.org/abs/2401.12540v1 | Chen Huang, Duanyu Feng, Wenqiang Lei, Jiancheng Lv | 2024-01-23T07:48:58Z | cs.IR, cs.CL | Deploying dense retrieval models efficiently is becoming increasingly
important across various industries. This is especially true for enterprise
search services, where customizing search engines to meet the time demands of
different enterprises in different domains is crucial. Motivated by this, we
develop a time-efficient approach called DREditor to edit the matching rule of
an off-the-shelf dense retrieval model to suit a specific domain. This is
achieved by directly calibrating the output embeddings of the model using an
efficient and effective linear mapping. This mapping is powered by an edit
operator that is obtained by solving a specially constructed least squares
problem. Compared to implicit rule modification via long-time finetuning, our
experimental results show that DREditor provides significant advantages on
different domain-specific datasets, dataset sources, retrieval models, and
computing devices. It consistently enhances time efficiency by 100-300 times
while maintaining comparable or even superior retrieval performance. In a
broader context, we take the first step to introduce a novel embedding
calibration approach for the retrieval task, filling the technical blank in the
current field of embedding calibration. This approach also paves the way for
building domain-specific dense retrieval models efficiently and inexpensively. | cs.IR | DREditor: An Time-efficient Approach for Building a Domain-specific Dense Retrieval Model |
2024-01-23T06:36:49Z | http://arxiv.org/pdf/2401.12522v2 | http://arxiv.org/abs/2401.12522v2 | Feng Lin, Hanling Yi, Hongbin Li, Yifan Yang, Xiaotian Yu, Guangming Lu, Rong Xiao | 2024-01-25T14:02:03Z | cs.CL, cs.AI, cs.LG | Large language models (LLMs) commonly employ autoregressive generation during
inference, leading to high memory bandwidth demand and consequently extended
latency. To mitigate this inefficiency, we present Bi-directional Tuning for
lossless Acceleration (BiTA), an innovative method expediting LLMs via
streamlined semi-autoregressive generation and draft verification. Inspired by
the concept of prompt tuning, we enhance LLMs with a parameter-efficient design
called bi-directional tuning for the capability in semi-autoregressive
generation. Employing efficient tree-based decoding, the models perform draft
candidate generation and verification in parallel, ensuring outputs identical
to their autoregressive counterparts under greedy sampling. BiTA serves as a
lightweight plug-in module, seamlessly boosting the inference efficiency of
existing LLMs without requiring additional assistance models or incurring
significant extra memory costs. Applying the proposed BiTA, LLaMA-2-70B-Chat
achieves a 2.7$\times$ speedup on the MT-Bench benchmark. Extensive experiments
confirm our method surpasses state-of-the-art acceleration techniques. | cs.CL | BiTA: Bi-Directional Tuning for Lossless Acceleration in Large Language Models |
2024-01-23T06:30:05Z | http://arxiv.org/pdf/2401.12520v1 | http://arxiv.org/abs/2401.12520v1 | Jiahui Zhao, Ziyi Meng, Stepan Gordeev, Zijie Pan, Dongjin Song, Sandro Steinbach, Caiwen Ding | 2024-01-23T06:30:05Z | cs.CL, cs.IR, cs.LG | With the rapid proliferation of textual data, predicting long texts has
emerged as a significant challenge in the domain of natural language
processing. Traditional text prediction methods encounter substantial
difficulties when grappling with long texts, primarily due to the presence of
redundant and irrelevant information, which impedes the model's capacity to
capture pivotal insights from the text. To address this issue, we introduce a
novel approach to long-text classification and prediction. Initially, we employ
embedding techniques to condense the long texts, aiming to diminish the
redundancy therein. Subsequently,the Bidirectional Encoder Representations from
Transformers (BERT) embedding method is utilized for text classification
training. Experimental outcomes indicate that our method realizes considerable
performance enhancements in classifying long texts of Preferential Trade
Agreements. Furthermore, the condensation of text through embedding methods not
only augments prediction accuracy but also substantially reduces computational
complexity. Overall, this paper presents a strategy for long-text prediction,
offering a valuable reference for researchers and engineers in the natural
language processing sphere. | cs.CL | Key Information Retrieval to Classify the Unstructured Data Content of Preferential Trade Agreements |
2024-01-23T05:20:35Z | http://arxiv.org/pdf/2401.12492v2 | http://arxiv.org/abs/2401.12492v2 | Nikita Soni, Niranjan Balasubramanian, H. Andrew Schwartz, Dirk Hovy | 2024-03-26T19:28:15Z | cs.CL, cs.AI, cs.LG | Incorporating human context into language models is the next frontier for
human-centered natural language processing. Currently, two pre-training methods
exist: group-wise attributes (e.g., over-45-year-olds) or individual traits.
Group attributes are coarse -- not all 45-year-olds write the same way -- while
modeling individual traits allows for a more personalized representation, but
requires more complex modeling and data. So far, it is unclear which
pre-training approach benefits what tasks. We compare pre-training models with
human context via 1) group attributes, 2) individual users, and 3) a combined
approach on 5 user- and document-level tasks. We find that pre-training with
both group and individual features significantly improves the two user-level
regression tasks like age estimation and personality assessment. Pre-training
on individual users significantly improves the three document-level
classification tasks like stance and topic detection. It even does well for
downstream tasks without historical user data. Our results suggest both
approaches have specific use cases, opening new avenues for human-centered
language modeling. | cs.CL | Comparing Pre-trained Human Language Models: Is it Better with Human Context as Groups, Individual Traits, or Both? |
2024-01-23T05:19:47Z | http://arxiv.org/pdf/2401.12491v1 | http://arxiv.org/abs/2401.12491v1 | Yunpu Zhao, Rui Zhang, Wenyi Li, Di Huang, Jiaming Guo, Shaohui Peng, Yifan Hao, Yuanbo Wen, Xing Hu, Zidong Du, Qi Guo, Ling Li, Yunji Chen | 2024-01-23T05:19:47Z | cs.CL, cs.AI | In the field of natural language processing, the rapid development of large
language model (LLM) has attracted more and more attention. LLMs have shown a
high level of creativity in various tasks, but the methods for assessing such
creativity are inadequate. The assessment of LLM creativity needs to consider
differences from humans, requiring multi-dimensional measurement while
balancing accuracy and efficiency. This paper aims to establish an efficient
framework for assessing the level of creativity in LLMs. By adapting the
modified Torrance Tests of Creative Thinking, the research evaluates the
creative performance of various LLMs across 7 tasks, emphasizing 4 criteria
including Fluency, Flexibility, Originality, and Elaboration. In this context,
we develop a comprehensive dataset of 700 questions for testing and an
LLM-based evaluation method. In addition, this study presents a novel analysis
of LLMs' responses to diverse prompts and role-play situations. We found that
the creativity of LLMs primarily falls short in originality, while excelling in
elaboration. Besides, the use of prompts and the role-play settings of the
model significantly influence creativity. Additionally, the experimental
results also indicate that collaboration among multiple LLMs can enhance
originality. Notably, our findings reveal a consensus between human evaluations
and LLMs regarding the personality traits that influence creativity. The
findings underscore the significant impact of LLM design on creativity and
bridges artificial intelligence and human creativity, offering insights into
LLMs' creativity and potential applications. | cs.CL | Assessing and Understanding Creativity in Large Language Models |
2024-01-23T03:56:22Z | http://arxiv.org/pdf/2401.12474v1 | http://arxiv.org/abs/2401.12474v1 | Keming Lu, Bowen Yu, Chang Zhou, Jingren Zhou | 2024-01-23T03:56:22Z | cs.CL, cs.LG | Considerable efforts have been invested in augmenting the role-playing
proficiency of open-source large language models (LLMs) by emulating
proprietary counterparts. Nevertheless, we posit that LLMs inherently harbor
role-play capabilities, owing to the extensive knowledge of characters and
potential dialogues ingrained in their vast training corpora. Thus, in this
study, we introduce Ditto, a self-alignment method for role-play. Ditto
capitalizes on character knowledge, encouraging an instruction-following LLM to
simulate role-play dialogues as a variant of reading comprehension. This method
creates a role-play training set comprising 4,000 characters, surpassing the
scale of currently available datasets by tenfold regarding the number of roles.
Subsequently, we fine-tune the LLM using this self-generated dataset to augment
its role-playing capabilities. Upon evaluating our meticulously constructed and
reproducible role-play benchmark and the roleplay subset of MT-Bench, Ditto, in
various parameter scales, consistently maintains a consistent role identity and
provides accurate role-specific knowledge in multi-turn role-play
conversations. Notably, it outperforms all open-source role-play baselines,
showcasing performance levels comparable to advanced proprietary chatbots.
Furthermore, we present the first comprehensive cross-supervision alignment
experiment in the role-play domain, revealing that the intrinsic capabilities
of LLMs confine the knowledge within role-play. Meanwhile, the role-play styles
can be easily acquired with the guidance of smaller models. We open-source
related resources at https://github.com/OFA-Sys/Ditto. | cs.CL | Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment |
2024-01-23T03:47:07Z | http://arxiv.org/pdf/2401.12472v1 | http://arxiv.org/abs/2401.12472v1 | Valerie Lim, Kai Wen Ng, Kenneth Lim | 2024-01-23T03:47:07Z | cs.CL | Natural Language Processing models like BERT can provide state-of-the-art
word embeddings for downstream NLP tasks. However, these models yet to perform
well on Semantic Textual Similarity, and may be too large to be deployed as
lightweight edge applications. We seek to apply a suitable contrastive learning
method based on the SimCSE paper, to a model architecture adapted from a
knowledge distillation based model, DistilBERT, to address these two issues.
Our final lightweight model DistilFace achieves an average of 72.1 in
Spearman's correlation on STS tasks, a 34.2 percent improvement over BERT base. | cs.CL | Contrastive Learning in Distilled Models |
2024-01-23T03:03:57Z | http://arxiv.org/pdf/2401.12461v1 | http://arxiv.org/abs/2401.12461v1 | Yichen Yang, Xin Liu, Kun He | 2024-01-23T03:03:57Z | cs.CL, cs.AI | Many adversarial defense methods have been proposed to enhance the
adversarial robustness of natural language processing models. However, most of
them introduce additional pre-set linguistic knowledge and assume that the
synonym candidates used by attackers are accessible, which is an ideal
assumption. We delve into adversarial training in the embedding space and
propose a Fast Adversarial Training (FAT) method to improve the model
robustness in the synonym-unaware scenario from the perspective of single-step
perturbation generation and perturbation initialization. Based on the
observation that the adversarial perturbations crafted by single-step and
multi-step gradient ascent are similar, FAT uses single-step gradient ascent to
craft adversarial examples in the embedding space to expedite the training
process. Based on the observation that the perturbations generated on the
identical training sample in successive epochs are similar, FAT fully utilizes
historical information when initializing the perturbation. Extensive
experiments demonstrate that FAT significantly boosts the robustness of BERT
models in the synonym-unaware scenario, and outperforms the defense baselines
under various attacks with character-level and word-level modifications. | cs.CL | Fast Adversarial Training against Textual Adversarial Attacks |
2024-01-23T01:33:09Z | http://arxiv.org/pdf/2401.12428v2 | http://arxiv.org/abs/2401.12428v2 | Songyun Qu, Shixin Zhao, Bing Li, Yintao He, Xuyi Cai, Lei Zhang, Ying Wang | 2024-05-08T06:44:41Z | cs.AR, cs.CL, D.3.4 | In recent years, various computing-in-memory (CIM) processors have been
presented, showing superior performance over traditional architectures. To
unleash the potential of various CIM architectures, such as device precision,
crossbar size, and crossbar number, it is necessary to develop compilation
tools that are fully aware of the CIM architectural details and implementation
diversity. However, due to the lack of architectural support in current popular
open-source compiling stacks, existing CIM designs either manually deploy
networks or build their own compilers, which is time-consuming and
labor-intensive. Although some works expose the specific CIM device programming
interfaces to compilers, they are often bound to a fixed CIM architecture,
lacking the flexibility to support the CIM architectures with different
computing granularity. On the other hand, existing compilation works usually
consider the scheduling of limited operation types (such as crossbar-bound
matrix-vector multiplication). Unlike conventional processors, CIM accelerators
are featured by their diverse architecture, circuit, and device, which cannot
be simply abstracted by a single level if we seek to fully explore the
advantages brought by CIM. Therefore, we propose CIM-MLC, a universal
multi-level compilation framework for general CIM architectures. We first
establish a general hardware abstraction for CIM architectures and computing
modes to represent various CIM accelerators. Based on the proposed abstraction,
CIM-MLC can compile tasks onto a wide range of CIM accelerators having
different devices, architectures, and programming interfaces. More importantly,
compared with existing compilation work, CIM-MLC can explore the mapping and
scheduling strategies across multiple architectural tiers, which form a
tractable yet effective design space, to achieve better scheduling and
instruction generation results. | cs.AR | CIM-MLC: A Multi-level Compilation Stack for Computing-In-Memory Accelerators |
2024-01-23T01:25:00Z | http://arxiv.org/pdf/2401.12425v3 | http://arxiv.org/abs/2401.12425v3 | Shubham Parashar, Zhiqiu Lin, Tian Liu, Xiangjue Dong, Yanan Li, Deva Ramanan, James Caverlee, Shu Kong | 2024-05-22T16:29:58Z | cs.CV, cs.CL, cs.LG | Vision-language models (VLMs) excel in zero-shot recognition but their
performance varies greatly across different visual concepts. For example,
although CLIP achieves impressive accuracy on ImageNet (60-80%), its
performance drops below 10% for more than ten concepts like night snake,
presumably due to their limited presence in the pretraining data. However,
measuring the frequency of concepts in VLMs' large-scale datasets is
challenging. We address this by using large language models (LLMs) to count the
number of pretraining texts that contain synonyms of these concepts. Our
analysis confirms that popular datasets, such as LAION, exhibit a long-tailed
concept distribution, yielding biased performance in VLMs. We also find that
downstream applications of VLMs, including visual chatbots (e.g., GPT-4V) and
text-to-image models (e.g., Stable Diffusion), often fail to recognize or
generate images of rare concepts identified by our method. To mitigate the
imbalanced performance of zero-shot VLMs, we propose REtrieval-Augmented
Learning (REAL). First, instead of prompting VLMs using the original class
names, REAL uses their most frequent synonyms found in pretraining texts. This
simple change already outperforms costly human-engineered and LLM-enriched
prompts over nine benchmark datasets. Second, REAL trains a linear classifier
on a small yet balanced set of pretraining data retrieved using concept
synonyms. REAL surpasses the previous zero-shot SOTA, using 400x less storage
and 10,000x less training time! | cs.CV | The Neglected Tails in Vision-Language Models |
2024-01-22T23:55:00Z | http://arxiv.org/pdf/2401.12413v2 | http://arxiv.org/abs/2401.12413v2 | Di Wu, Shaomu Tan, Yan Meng, David Stap, Christof Monz | 2024-02-26T23:23:31Z | cs.CL, cs.LG | Zero-shot translation aims to translate between language pairs not seen
during training in Multilingual Machine Translation (MMT) and is largely
considered an open problem. A common, albeit resource-consuming, solution is to
add as many related translation directions as possible to the training corpus.
In this paper, we show that for an English-centric model, surprisingly large
zero-shot improvements can be achieved by simply fine-tuning with a very small
amount of multi-parallel data. For example, on the EC30 dataset, we obtain up
to +21.7 ChrF non-English overall improvements (870 directions) by using only
100 multi-parallel samples while preserving English-centric translation
quality. When investigating the size effect of fine-tuning data and its
transfer capabilities, we found that already a small, randomly sampled set of
fine-tuning directions is sufficient to achieve comparable improvements. The
resulting non-English performance is close to the complete translation upper
bound. Even in a minimal setting -- fine-tuning with only one single sample --
the well-known off-target issue is almost completely resolved, explaining parts
-- but not all -- of the observed improvements in translation quality. | cs.CL | How Far Can 100 Samples Go? Unlocking Overall Zero-Shot Multilingual Translation via Tiny Multi-Parallel Data |
2024-01-22T23:35:09Z | http://arxiv.org/pdf/2401.12406v1 | http://arxiv.org/abs/2401.12406v1 | Momin Abbas, Yi Zhou, Parikshit Ram, Nathalie Baracaldo, Horst Samulowitz, Theodoros Salonidis, Tianyi Chen | 2024-01-22T23:35:09Z | cs.CL, cs.AI, cs.LG | In-context learning (ICL) is a new paradigm for natural language processing
that utilizes Generative Pre-trained Transformer (GPT)-like models. This
approach uses prompts that include in-context demonstrations to generate the
corresponding output for a new query input. However, applying ICL in real cases
does not scale with the number of samples, and lacks robustness to different
prompt templates and demonstration permutations. In this paper, we first show
that GPT-like models using ICL result in unreliable predictions based on a new
metric based on Shannon entropy. Then, to solve this problem, we propose a new
technique called the Linear Probe Calibration (LinC), a method that calibrates
the model's output probabilities, resulting in reliable predictions and
improved performance, while requiring only minimal additional samples (as few
as five labeled data samples). LinC significantly enhances the ICL test
performance of GPT models on various benchmark datasets, with an average
improvement of up to 21%, and up to a 50% improvement in some cases, and
significantly boosts the performance of PEFT methods, especially in the low
resource regime. Moreover, LinC achieves lower expected calibration error, and
is highly robust to varying label proportions, prompt templates, and
demonstration permutations. Our code is available at
\url{https://github.com/mominabbass/LinC}. | cs.CL | Enhancing In-context Learning via Linear Probe Calibration |
2024-01-22T22:16:55Z | http://arxiv.org/pdf/2401.12382v1 | http://arxiv.org/abs/2401.12382v1 | Fabian Nwaoha, Ziyad Gaffar, Ho Joon Chun, Marina Sokolova | 2024-01-22T22:16:55Z | cs.CL, cs.LG, cs.SI, I.2.6 | We report results of a longitudinal sentiment classification of Reddit posts
written by students of four major Canadian universities. We work with the texts
of the posts, concentrating on the years 2020-2023. By finely tuning a
sentiment threshold to a range of [-0.075,0.075], we successfully built
classifiers proficient in categorizing post sentiments into positive and
negative categories. Noticeably, our sentiment classification results are
consistent across the four university data sets. | cs.CL | Longitudinal Sentiment Classification of Reddit Posts |
2024-01-22T21:59:00Z | http://arxiv.org/pdf/2401.12375v1 | http://arxiv.org/abs/2401.12375v1 | Tubo Faustinah Nemieboka, Ikechukwu E. Onyenwe, Doris C. Asogwa | 2024-01-22T21:59:00Z | cs.CL, cs.AI | In recent years, advancements in Natural Language Processing (NLP) techniques
have revolutionized the field of accessibility and exclusivity of testing,
particularly for visually impaired students (VIS). CBT has shown in years back
its relevance in terms of administering exams electronically, making the test
process easier, providing quicker and more accurate results, and offering
greater flexibility and accessibility for candidates. Yet, its relevance was
not felt by the visually impaired students as they cannot access printed
documents. Hence, in this paper, we present an NLP-driven Computer-Based Test
guide for visually impaired students. It employs a speech technology
pre-trained methods to provide real-time assistance and support to visually
impaired students. The system utilizes NLP technologies to convert the
text-based questions and the associated options in a machine-readable format.
Subsequently, the speech technology pre-trained model processes the converted
text enabling the VIS to comprehend and analyze the content. Furthermore, we
validated that this pre-trained model is not perverse by testing for accuracy
using sample audio datasets labels (A, B, C, D, E, F, G) to compare with the
voice recordings obtained from 20 VIS which is been predicted by the system to
attain values for precision, recall, and F1-scores. These metrics are used to
assess the performance of the pre-trained model and have indicated that it is
proficient enough to give its better performance to the evaluated system. The
methodology adopted for this system is Object Oriented Analysis and Design
Methodology (OOADM) where Objects are discussed and built by modeling
real-world instances. | cs.CL | Development of an NLP-driven computer-based test guide for visually impaired students |
2024-01-22T20:17:06Z | http://arxiv.org/pdf/2401.12343v1 | http://arxiv.org/abs/2401.12343v1 | Hanchen Ye, David Z. Pan, Chris Leary, Deming Chen, Xiaoqing Xu | 2024-01-22T20:17:06Z | cs.CL | This paper proposes ISDC, a novel feedback-guided iterative system of
difference constraints (SDC) scheduling algorithm for high-level synthesis
(HLS). ISDC leverages subgraph extraction-based low-level feedback from
downstream tools like logic synthesizers to iteratively refine HLS scheduling.
Technical innovations include: (1) An enhanced SDC formulation that effectively
integrates low-level feedback into the linear-programming (LP) problem; (2) A
fanout and window-based subgraph extraction mechanism driving the feedback
cycle; (3) A no-human-in-loop ISDC flow compatible with a wide range of
downstream tools and process design kits (PDKs). Evaluation shows that ISDC
reduces register usage by 28.5% against an industrial-strength open-source HLS
tool. | cs.CL | Subgraph Extraction-based Feedback-guided Iterative Scheduling for HLS |
2024-01-22T19:39:05Z | http://arxiv.org/pdf/2401.12326v1 | http://arxiv.org/abs/2401.12326v1 | Feng Xiong, Thanet Markchom, Ziwei Zheng, Subin Jung, Varun Ojha, Huizhi Liang | 2024-01-22T19:39:05Z | cs.CL, cs.AI | SemEval-2024 Task 8 introduces the challenge of identifying machine-generated
texts from diverse Large Language Models (LLMs) in various languages and
domains. The task comprises three subtasks: binary classification in
monolingual and multilingual (Subtask A), multi-class classification (Subtask
B), and mixed text detection (Subtask C). This paper focuses on Subtask A & B.
Each subtask is supported by three datasets for training, development, and
testing. To tackle this task, two methods: 1) using traditional machine
learning (ML) with natural language preprocessing (NLP) for feature extraction,
and 2) fine-tuning LLMs for text classification. The results show that
transformer models, particularly LoRA-RoBERTa, exceed traditional ML methods in
effectiveness, with majority voting being particularly effective in
multilingual contexts for identifying machine-generated texts. | cs.CL | Fine-tuning Large Language Models for Multigenerator, Multidomain, and Multilingual Machine-Generated Text Detection |
2024-01-22T19:00:11Z | http://arxiv.org/pdf/2401.12295v1 | http://arxiv.org/abs/2401.12295v1 | Leonardo Castro-Gonzalez, Yi-Ling Chung, Hannak Rose Kirk, John Francis, Angus R. Williams, Pica Johansson, Jonathan Bright | 2024-01-22T19:00:11Z | cs.CL, I.2.7; J.4 | The field of machine learning has recently made significant progress in
reducing the requirements for labelled training data when building new models.
These `cheaper' learning techniques hold significant potential for the social
sciences, where development of large labelled training datasets is often a
significant practical impediment to the use of machine learning for analytical
tasks. In this article we review three `cheap' techniques that have developed
in recent years: weak supervision, transfer learning and prompt engineering.
For the latter, we also review the particular case of zero-shot prompting of
large language models. For each technique we provide a guide of how it works
and demonstrate its application across six different realistic social science
applications (two different tasks paired with three different dataset makeups).
We show good performance for all techniques, and in particular we demonstrate
how prompting of large language models can achieve high accuracy at very low
cost. Our results are accompanied by a code repository to make it easy for
others to duplicate our work and use it in their own research. Overall, our
article is intended to stimulate further uptake of these techniques in the
social sciences. | cs.CL | Cheap Learning: Maximising Performance of Language Models for Social Data Science Using Minimal Data |
2024-01-22T19:00:08Z | http://arxiv.org/pdf/2401.12292v2 | http://arxiv.org/abs/2401.12292v2 | Weixin Chen, Dawn Song, Bo Li | 2024-01-31T06:44:42Z | cs.CL, cs.AI | Truthfulness is paramount for large language models (LLMs) as they are
increasingly deployed in real-world applications. However, existing LLMs still
struggle with generating truthful content, as evidenced by their modest
performance on benchmarks like TruthfulQA. To address this issue, we propose
GRAdual self-truTHifying (GRATH), a novel post-processing method to enhance
truthfulness of LLMs. GRATH utilizes out-of-domain question prompts to generate
pairwise truthfulness training data with each pair containing a question and
its correct and incorrect answers, and then optimizes the model via direct
preference optimization (DPO) to learn from the truthfulness difference between
answer pairs. GRATH iteratively refines truthfulness data and updates the
model, leading to a gradual improvement in model truthfulness in a
self-supervised manner. Empirically, we evaluate GRATH using different 7B-LLMs
and compare with LLMs with similar or even larger sizes on benchmark datasets.
Our results show that GRATH effectively improves LLMs' truthfulness without
compromising other core capabilities. Notably, GRATH achieves state-of-the-art
performance on TruthfulQA, with MC1 accuracy of 54.71% and MC2 accuracy of
69.10%, which even surpass those on 70B-LLMs. | cs.CL | GRATH: Gradual Self-Truthifying for Large Language Models |
2024-01-22T18:51:07Z | http://arxiv.org/pdf/2401.12208v1 | http://arxiv.org/abs/2401.12208v1 | Zhihong Chen, Maya Varma, Jean-Benoit Delbrouck, Magdalini Paschali, Louis Blankemeier, Dave Van Veen, Jeya Maria Jose Valanarasu, Alaa Youssef, Joseph Paul Cohen, Eduardo Pontes Reis, Emily B. Tsai, Andrew Johnston, Cameron Olsen, Tanishq Mathew Abraham, Sergios Gatidis, Akshay S. Chaudhari, Curtis Langlotz | 2024-01-22T18:51:07Z | cs.CV, cs.CL | Chest X-rays (CXRs) are the most frequently performed imaging test in
clinical practice. Recent advances in the development of vision-language
foundation models (FMs) give rise to the possibility of performing automated
CXR interpretation, which can assist physicians with clinical decision-making
and improve patient outcomes. However, developing FMs that can accurately
interpret CXRs is challenging due to the (1) limited availability of
large-scale vision-language datasets in the medical image domain, (2) lack of
vision and language encoders that can capture the complexities of medical data,
and (3) absence of evaluation frameworks for benchmarking the abilities of FMs
on CXR interpretation. In this work, we address these challenges by first
introducing \emph{CheXinstruct} - a large-scale instruction-tuning dataset
curated from 28 publicly-available datasets. We then present \emph{CheXagent} -
an instruction-tuned FM capable of analyzing and summarizing CXRs. To build
CheXagent, we design a clinical large language model (LLM) for parsing
radiology reports, a vision encoder for representing CXR images, and a network
to bridge the vision and language modalities. Finally, we introduce
\emph{CheXbench} - a novel benchmark designed to systematically evaluate FMs
across 8 clinically-relevant CXR interpretation tasks. Extensive quantitative
evaluations and qualitative reviews with five expert radiologists demonstrate
that CheXagent outperforms previously-developed general- and medical-domain FMs
on CheXbench tasks. Furthermore, in an effort to improve model transparency, we
perform a fairness evaluation across factors of sex, race and age to highlight
potential performance disparities. Our project is at
\url{https://stanford-aimi.github.io/chexagent.html}. | cs.CV | CheXagent: Towards a Foundation Model for Chest X-Ray Interpretation |
2024-01-22T18:39:40Z | http://arxiv.org/pdf/2401.12200v2 | http://arxiv.org/abs/2401.12200v2 | Bowen Zhao, Hannaneh Hajishirzi, Qingqing Cao | 2024-06-04T06:39:23Z | cs.CL, cs.LG | Fine-tuning and inference with large Language Models (LM) are generally known
to be expensive. Parameter-efficient fine-tuning over pretrained LMs reduces
training memory by updating a small number of LM parameters but does not
improve inference efficiency. Structured pruning improves LM inference
efficiency by removing consistent parameter blocks, yet often increases
training memory and time. To improve both training and inference efficiency, we
introduce APT that adaptively prunes and tunes parameters for the LMs. At the
early stage of fine-tuning, APT dynamically adds salient tuning parameters for
fast and accurate convergence while discarding unimportant parameters for
efficiency. Compared to baselines, our experiments show that APT maintains up
to 98% task performance when pruning RoBERTa and T5 models with 40% parameters
left while keeping 86.4% LLaMA models' performance with 70% parameters
remained. Furthermore, APT speeds up LMs fine-tuning by up to 8x and reduces
large LMs memory training footprint by up to 70%. | cs.CL | APT: Adaptive Pruning and Tuning Pretrained Language Models for Efficient Training and Inference |
2024-01-22T18:34:42Z | http://arxiv.org/pdf/2401.12192v4 | http://arxiv.org/abs/2401.12192v4 | Yiyi Chen, Heather Lent, Johannes Bjerva | 2024-06-05T10:22:00Z | cs.CL, cs.AI, cs.CR | Textual data is often represented as real-numbered embeddings in NLP,
particularly with the popularity of large language models (LLMs) and Embeddings
as a Service (EaaS). However, storing sensitive information as embeddings can
be susceptible to security breaches, as research shows that text can be
reconstructed from embeddings, even without knowledge of the underlying model.
While defence mechanisms have been explored, these are exclusively focused on
English, leaving other languages potentially exposed to attacks. This work
explores LLM security through multilingual embedding inversion. We define the
problem of black-box multilingual and cross-lingual inversion attacks, and
explore their potential implications. Our findings suggest that multilingual
LLMs may be more vulnerable to inversion attacks, in part because English-based
defences may be ineffective. To alleviate this, we propose a simple masking
defense effective for both monolingual and multilingual models. This study is
the first to investigate multilingual inversion attacks, shedding light on the
differences in attacks and defenses across monolingual and multilingual
settings. | cs.CL | Text Embedding Inversion Security for Multilingual Language Models |
2024-01-22T18:27:08Z | http://arxiv.org/pdf/2401.12187v1 | http://arxiv.org/abs/2401.12187v1 | Alexandre Ramé, Nino Vieillard, Léonard Hussenot, Robert Dadashi, Geoffrey Cideron, Olivier Bachem, Johan Ferret | 2024-01-22T18:27:08Z | cs.LG, cs.AI, cs.CL | Aligning large language models (LLMs) with human preferences through
reinforcement learning (RLHF) can lead to reward hacking, where LLMs exploit
failures in the reward model (RM) to achieve seemingly high rewards without
meeting the underlying objectives. We identify two primary challenges when
designing RMs to mitigate reward hacking: distribution shifts during the RL
process and inconsistencies in human preferences. As a solution, we propose
Weight Averaged Reward Models (WARM), first fine-tuning multiple RMs, then
averaging them in the weight space. This strategy follows the observation that
fine-tuned weights remain linearly mode connected when sharing the same
pre-training. By averaging weights, WARM improves efficiency compared to the
traditional ensembling of predictions, while improving reliability under
distribution shifts and robustness to preference inconsistencies. Our
experiments on summarization tasks, using best-of-N and RL methods, shows that
WARM improves the overall quality and alignment of LLM predictions; for
example, a policy RL fine-tuned with WARM has a 79.4% win rate against a policy
RL fine-tuned with a single RM. | cs.LG | WARM: On the Benefits of Weight Averaged Reward Models |
2024-01-22T18:11:01Z | http://arxiv.org/pdf/2401.12181v1 | http://arxiv.org/abs/2401.12181v1 | Wes Gurnee, Theo Horsley, Zifan Carl Guo, Tara Rezaei Kheirkhah, Qinyi Sun, Will Hathaway, Neel Nanda, Dimitris Bertsimas | 2024-01-22T18:11:01Z | cs.LG, cs.AI, cs.CL | A basic question within the emerging field of mechanistic interpretability is
the degree to which neural networks learn the same underlying mechanisms. In
other words, are neural mechanisms universal across different models? In this
work, we study the universality of individual neurons across GPT2 models
trained from different initial random seeds, motivated by the hypothesis that
universal neurons are likely to be interpretable. In particular, we compute
pairwise correlations of neuron activations over 100 million tokens for every
neuron pair across five different seeds and find that 1-5\% of neurons are
universal, that is, pairs of neurons which consistently activate on the same
inputs. We then study these universal neurons in detail, finding that they
usually have clear interpretations and taxonomize them into a small number of
neuron families. We conclude by studying patterns in neuron weights to
establish several universal functional roles of neurons in simple circuits:
deactivating attention heads, changing the entropy of the next token
distribution, and predicting the next token to (not) be within a particular
set. | cs.LG | Universal Neurons in GPT2 Language Models |
2024-01-22T18:09:52Z | http://arxiv.org/pdf/2401.12178v1 | http://arxiv.org/abs/2401.12178v1 | Karel D'Oosterlinck, Omar Khattab, François Remy, Thomas Demeester, Chris Develder, Christopher Potts | 2024-01-22T18:09:52Z | cs.CL, cs.AI | Multi-label classification problems with thousands of classes are hard to
solve with in-context learning alone, as language models (LMs) might lack prior
knowledge about the precise classes or how to assign them, and it is generally
infeasible to demonstrate every class in a prompt. We propose a general
program, $\texttt{Infer--Retrieve--Rank}$, that defines multi-step interactions
between LMs and retrievers to efficiently tackle such problems. We implement
this program using the $\texttt{DSPy}$ programming model, which specifies
in-context systems in a declarative manner, and use $\texttt{DSPy}$ optimizers
to tune it towards specific datasets by bootstrapping only tens of few-shot
examples. Our primary extreme classification program, optimized separately for
each task, attains state-of-the-art results across three benchmarks (HOUSE,
TECH, TECHWOLF). We apply the same program to a benchmark with vastly different
characteristics and attain competitive performance as well (BioDEX). Unlike
prior work, our proposed solution requires no finetuning, is easily applicable
to new tasks, alleviates prompt engineering, and requires only tens of labeled
examples. Our code is public at https://github.com/KarelDO/xmc.dspy. | cs.CL | In-Context Learning for Extreme Multi-Label Classification |
2024-01-22T18:01:01Z | http://arxiv.org/pdf/2401.12168v1 | http://arxiv.org/abs/2401.12168v1 | Boyuan Chen, Zhuo Xu, Sean Kirmani, Brian Ichter, Danny Driess, Pete Florence, Dorsa Sadigh, Leonidas Guibas, Fei Xia | 2024-01-22T18:01:01Z | cs.CV, cs.CL, cs.LG, cs.RO | Understanding and reasoning about spatial relationships is a fundamental
capability for Visual Question Answering (VQA) and robotics. While Vision
Language Models (VLM) have demonstrated remarkable performance in certain VQA
benchmarks, they still lack capabilities in 3D spatial reasoning, such as
recognizing quantitative relationships of physical objects like distances or
size differences. We hypothesize that VLMs' limited spatial reasoning
capability is due to the lack of 3D spatial knowledge in training data and aim
to solve this problem by training VLMs with Internet-scale spatial reasoning
data. To this end, we present a system to facilitate this approach. We first
develop an automatic 3D spatial VQA data generation framework that scales up to
2 billion VQA examples on 10 million real-world images. We then investigate
various factors in the training recipe, including data quality, training
pipeline, and VLM architecture. Our work features the first internet-scale 3D
spatial reasoning dataset in metric space. By training a VLM on such data, we
significantly enhance its ability on both qualitative and quantitative spatial
VQA. Finally, we demonstrate that this VLM unlocks novel downstream
applications in chain-of-thought spatial reasoning and robotics due to its
quantitative estimation capability. Project website:
https://spatial-vlm.github.io/ | cs.CV | SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities |
2024-01-22T17:26:55Z | http://arxiv.org/pdf/2401.12143v2 | http://arxiv.org/abs/2401.12143v2 | Nathan Godey, Éric de la Clergerie, Benoît Sagot | 2024-01-24T16:07:00Z | cs.CL | The representation degeneration problem is a phenomenon that is widely
observed among self-supervised learning methods based on Transformers. In NLP,
it takes the form of anisotropy, a singular property of hidden representations
which makes them unexpectedly close to each other in terms of angular distance
(cosine-similarity). Some recent works tend to show that anisotropy is a
consequence of optimizing the cross-entropy loss on long-tailed distributions
of tokens. We show in this paper that anisotropy can also be observed
empirically in language models with specific objectives that should not suffer
directly from the same consequences. We also show that the anisotropy problem
extends to Transformers trained on other modalities. Our observations suggest
that anisotropy is actually inherent to Transformers-based models. | cs.CL | Anisotropy Is Inherent to Self-Attention in Transformers |
2024-01-22T17:11:37Z | http://arxiv.org/pdf/2401.12273v1 | http://arxiv.org/abs/2401.12273v1 | Ashutosh Kumar, Sagarika Singh, Shiv Vignesh Murty, Swathy Ragupathy | 2024-01-22T17:11:37Z | cs.CR, cs.AI, cs.CL | This paper comprehensively explores the ethical challenges arising from
security threats to Language Learning Models (LLMs). These intricate digital
repositories are increasingly integrated into our daily lives, making them
prime targets for attacks that can compromise their training data and the
confidentiality of their data sources. The paper delves into the nuanced
ethical repercussions of such security threats on society and individual
privacy. We scrutinize five major threats: prompt injection, jailbreaking,
Personal Identifiable Information (PII) exposure, sexually explicit content,
and hate based content, going beyond mere identification to assess their
critical ethical consequences and the urgency they create for robust defensive
strategies. The escalating reliance on LLMs underscores the crucial need for
ensuring these systems operate within the bounds of ethical norms, particularly
as their misuse can lead to significant societal and individual harm. We
propose conceptualizing and developing an evaluative tool tailored for LLMs,
which would serve a dual purpose, guiding developers and designers in
preemptive fortification of backend systems and scrutinizing the ethical
dimensions of LLM chatbot responses during the testing phase. By comparing LLM
responses with those expected from humans in a moral context, we aim to discern
the degree to which AI behaviors align with the ethical values held by a
broader society. Ultimately, this paper not only underscores the ethical
troubles presented by LLMs, it also highlights a path toward cultivating trust
in these systems. | cs.CR | The Ethics of Interaction: Mitigating Security Threats in LLMs |
2024-01-22T16:57:05Z | http://arxiv.org/pdf/2401.12117v2 | http://arxiv.org/abs/2401.12117v2 | Kian Ahrabian, Zhivar Sourati, Kexuan Sun, Jiarui Zhang, Yifan Jiang, Fred Morstatter, Jay Pujara | 2024-02-13T07:04:30Z | cs.CL | While large language models (LLMs) are still being adopted to new domains and
utilized in novel applications, we are experiencing an influx of the new
generation of foundation models, namely multi-modal large language models
(MLLMs). These models integrate verbal and visual information, opening new
possibilities to demonstrate more complex reasoning abilities at the
intersection of the two modalities. However, despite the revolutionizing
prospect of MLLMs, our understanding of their reasoning abilities is limited.
In this study, we assess the nonverbal abstract reasoning abilities of
open-source and closed-source MLLMs using variations of Raven's Progressive
Matrices. Our experiments expose the difficulty of solving such problems while
showcasing the immense gap between open-source and closed-source models. We
also reveal critical shortcomings with individual visual and textual modules,
subjecting the models to low-performance ceilings. Finally, to improve MLLMs'
performance, we experiment with various methods, such as Chain-of-Thought
prompting, resulting in a significant (up to 100%) boost in performance. | cs.CL | The Curious Case of Nonverbal Abstract Reasoning with Multi-Modal Large Language Models |
2024-01-22T16:35:00Z | http://arxiv.org/pdf/2401.12097v3 | http://arxiv.org/abs/2401.12097v3 | Pranjal A. Chitale, Jay Gala, Raj Dabre | 2024-06-04T19:37:52Z | cs.CL | Recent interest has surged in employing Large Language Models (LLMs) for
machine translation (MT) via in-context learning (ICL) (Vilar et al., 2023).
Most prior studies primarily focus on optimizing translation quality, with
limited attention to understanding the specific aspects of ICL that influence
the said quality. To this end, we perform the first of its kind, an exhaustive
study of in-context learning for machine translation. We first establish that
ICL is primarily example-driven and not instruction-driven. Following this, we
conduct an extensive exploration of various aspects of the examples to
understand their influence on downstream performance. Our analysis includes
factors such as quality and quantity of demonstrations, spatial proximity, and
source versus target originality. Further, we also investigate challenging
scenarios involving indirectness and misalignment of examples to understand the
limits of ICL. While we establish the significance of the quality of the target
distribution over the source distribution of demonstrations, we further observe
that perturbations sometimes act as regularizers, resulting in performance
improvements. Surprisingly, ICL does not necessitate examples from the same
task, and a related task with the same target distribution proves sufficient.
We hope that our study acts as a guiding resource for considerations in
utilizing ICL for MT. Our code is available on
https://github.com/PranjalChitale/in-context-mt-analysis. | cs.CL | An Empirical Study of In-context Learning in LLMs for Machine Translation |
2024-01-22T16:25:47Z | http://arxiv.org/pdf/2401.12088v1 | http://arxiv.org/abs/2401.12088v1 | Aissatou Diallo, Antonis Bikakis, Luke Dickens, Anthony Hunter, Rob Miller | 2024-01-22T16:25:47Z | cs.CL | Cooking recipes are one of the most readily available kinds of procedural
text. They consist of natural language instructions that can be challenging to
interpret. In this paper, we propose a model to identify relevant information
from recipes and generate a graph to represent the sequence of actions in the
recipe. In contrast with other approaches, we use an unsupervised approach. We
iteratively learn the graph structure and the parameters of a $\mathsf{GNN}$
encoding the texts (text-to-graph) one sequence at a time while providing the
supervision by decoding the graph into text (graph-to-text) and comparing the
generated text to the input. We evaluate the approach by comparing the
identified entities with annotated datasets, comparing the difference between
the input and output texts, and comparing our generated graphs with those
generated by state of the art methods. | cs.CL | Unsupervised Learning of Graph from Recipes |
2024-01-22T16:25:27Z | http://arxiv.org/pdf/2401.12087v2 | http://arxiv.org/abs/2401.12087v2 | Keqin Peng, Liang Ding, Yancheng Yuan, Xuebo Liu, Min Zhang, Yuanxin Ouyang, Dacheng Tao | 2024-06-23T13:45:14Z | cs.CL | Large language models (LLMs) have shown an impressive ability to perform a
wide range of tasks using in-context learning (ICL), where a few examples are
used to describe a task to the model. However, the performance of ICL varies
significantly with the choice of demonstrations, and it is still unclear why
this happens or what factors will influence its choice. In this work, we first
revisit the factors contributing to this variance from both data and model
aspects, and find that the choice of demonstration is both data- and
model-dependent. We further proposed a data- and model-dependent demonstration
selection method, \textbf{TopK + ConE}, based on the assumption that
\textit{the performance of a demonstration positively correlates with its
contribution to the model's understanding of the test samples}, resulting in a
simple and effective recipe for ICL. Empirically, our method yields consistent
improvements in both language understanding and generation tasks with different
model scales. Further analyses confirm that, besides the generality and
stability under different circumstances, our method provides a unified
explanation for the effectiveness of previous methods. Code will be released. | cs.CL | Revisiting Demonstration Selection Strategies in In-Context Learning |
2024-01-22T16:24:43Z | http://arxiv.org/pdf/2401.12086v1 | http://arxiv.org/abs/2401.12086v1 | Alizée Pace, Jonathan Mallinson, Eric Malmi, Sebastian Krause, Aliaksei Severyn | 2024-01-22T16:24:43Z | cs.CL, cs.AI, cs.LG | The success of reinforcement learning from human feedback (RLHF) in language
model alignment is strongly dependent on the quality of the underlying reward
model. In this paper, we present a novel approach to improve reward model
quality by generating synthetic preference data, thereby augmenting the
training dataset with on-policy, high-quality preference pairs. Motivated by
the promising results of Best-of-N sampling strategies in language model
training, we extend their application to reward model training. This results in
a self-training strategy to generate preference pairs by selecting the best and
worst candidates in a pool of responses to a given query. Empirically, we find
that this approach improves the performance of any reward model, with an effect
comparable to the addition of a similar quantity of human preference data. This
work opens up new avenues of research for improving RLHF for language model
alignment, by offering synthetic preference generation as a solution to reward
modeling challenges. | cs.CL | West-of-N: Synthetic Preference Generation for Improved Reward Modeling |
2024-01-22T16:20:14Z | http://arxiv.org/pdf/2401.12078v1 | http://arxiv.org/abs/2401.12078v1 | Jonas Wallat, Adam Jatowt, Avishek Anand | 2024-01-22T16:20:14Z | cs.CL | Large language models (LLMs) have recently gained significant attention due
to their unparalleled ability to perform various natural language processing
tasks. These models, benefiting from their advanced natural language
understanding capabilities, have demonstrated impressive zero-shot performance.
However, the pre-training data utilized in LLMs is often confined to a specific
corpus, resulting in inherent freshness and temporal scope limitations.
Consequently, this raises concerns regarding the effectiveness of LLMs for
tasks involving temporal intents. In this study, we aim to investigate the
underlying limitations of general-purpose LLMs when deployed for tasks that
require a temporal understanding. We pay particular attention to handling
factual temporal knowledge through three popular temporal QA datasets.
Specifically, we observe low performance on detailed questions about the past
and, surprisingly, for rather new information. In manual and automatic testing,
we find multiple temporal errors and characterize the conditions under which QA
performance deteriorates. Our analysis contributes to understanding LLM
limitations and offers valuable insights into developing future models that can
better cater to the demands of temporally-oriented tasks. The code is
available\footnote{https://github.com/jwallat/temporalblindspots}. | cs.CL | Temporal Blind Spots in Large Language Models |
2024-01-22T16:13:45Z | http://arxiv.org/pdf/2401.12072v1 | http://arxiv.org/abs/2401.12072v1 | Fadli Aulawi Al Ghiffari, Ika Alfina, Kurniawati Azizah | 2024-01-22T16:13:45Z | cs.CL | While structure learning achieves remarkable performance in high-resource
languages, the situation differs for under-represented languages due to the
scarcity of annotated data. This study focuses on assessing the efficacy of
transfer learning in enhancing dependency parsing for Javanese, a language
spoken by 80 million individuals but characterized by limited representation in
natural language processing. We utilized the Universal Dependencies dataset
consisting of dependency treebanks from more than 100 languages, including
Javanese. We propose two learning strategies to train the model: transfer
learning (TL) and hierarchical transfer learning (HTL). While TL only uses a
source language to pre-train the model, the HTL method uses a source language
and an intermediate language in the learning process. The results show that our
best model uses the HTL method, which improves performance with an increase of
10% for both UAS and LAS evaluations compared to the baseline model. | cs.CL | Cross-lingual Transfer Learning for Javanese Dependency Parsing |
2024-01-22T16:09:47Z | http://arxiv.org/pdf/2401.12070v2 | http://arxiv.org/abs/2401.12070v2 | Abhimanyu Hans, Avi Schwarzschild, Valeriia Cherepanova, Hamid Kazemi, Aniruddha Saha, Micah Goldblum, Jonas Geiping, Tom Goldstein | 2024-07-01T15:17:10Z | cs.CL, cs.AI, cs.LG | Detecting text generated by modern large language models is thought to be
hard, as both LLMs and humans can exhibit a wide range of complex behaviors.
However, we find that a score based on contrasting two closely related language
models is highly accurate at separating human-generated and machine-generated
text. Based on this mechanism, we propose a novel LLM detector that only
requires simple calculations using a pair of pre-trained LLMs. The method,
called Binoculars, achieves state-of-the-art accuracy without any training
data. It is capable of spotting machine text from a range of modern LLMs
without any model-specific modifications. We comprehensively evaluate
Binoculars on a number of text sources and in varied situations. Over a wide
range of document types, Binoculars detects over 90% of generated samples from
ChatGPT (and other LLMs) at a false positive rate of 0.01%, despite not being
trained on any ChatGPT data. | cs.CL | Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text |
2024-01-22T14:53:59Z | http://arxiv.org/pdf/2401.12005v2 | http://arxiv.org/abs/2401.12005v2 | Weihang Huang, Akira Murakami, Jack Grieve | 2024-02-13T04:18:45Z | cs.CL | In this paper, we introduce an authorship attribution method called Authorial
Language Models (ALMs) that involves identifying the most likely author of a
questioned document based on the perplexity of the questioned document
calculated for a set of causal language models fine-tuned on the writings of a
set of candidate author. We benchmarked ALMs against state-of-art-systems using
the CCAT50 dataset and the Blogs50 datasets. We find that ALMs achieves a
macro-average accuracy score of 83.6% on Blogs50, outperforming all other
methods, and 74.9% on CCAT50, matching the performance of the best method. To
assess the performance of ALMs on shorter texts, we also conducted text
ablation testing. We found that to reach a macro-average accuracy of 70%, ALMs
needs 40 tokens on Blogs50 and 400 tokens on CCAT50, while to reach 60% ALMs
requires 20 tokens on Blogs50 and 70 tokens on CCAT50. | cs.CL | ALMs: Authorial Language Models for Authorship Attribution |
2024-01-22T14:48:13Z | http://arxiv.org/pdf/2402.07907v1 | http://arxiv.org/abs/2402.07907v1 | Dimitrios Sidiropoulos, Christos-Nikolaos Anagnostopoulos | 2024-01-22T14:48:13Z | cs.HC, cs.CL | Artificial Intelligence (AI) in recent years has shown an unprecedentedly
impressive development, tending to play a catalytic role in all aspects of
life. The interest of the academic community, but also of governments, is huge
in the dynamics of AI and is reflected by the truly explosive amount of
investment and research that is underway. Enthusiastic opinions and statements
about AI are made every day, but at the same time they also bring to the fore
alarming predictions about its effects. This paper aims to describe the
opportunities emerging from the use of artificial intelligence and ChatGPT to
improve education, but also to identify the challenges and ethical issues that
arise. | cs.HC | Applications, challenges and ethical issues of AI and ChatGPT in education |
2024-01-22T14:24:03Z | http://arxiv.org/pdf/2401.11972v2 | http://arxiv.org/abs/2401.11972v2 | Rrubaa Panchendrarajan, Arkaitz Zubiaga | 2024-03-18T17:05:30Z | cs.CL | The advancement of machine learning and symbolic approaches have underscored
their strengths and weaknesses in Natural Language Processing (NLP). While
machine learning approaches are powerful in identifying patterns in data, they
often fall short in learning commonsense and the factual knowledge required for
the NLP tasks. Meanwhile, the symbolic methods excel in representing
knowledge-rich data. However, they struggle to adapt dynamic data and
generalize the knowledge. Bridging these two paradigms through hybrid
approaches enables the alleviation of weaknesses in both while preserving their
strengths. Recent studies extol the virtues of this union, showcasing promising
results in a wide range of NLP tasks. In this paper, we present an overview of
hybrid approaches used for NLP. Specifically, we delve into the
state-of-the-art hybrid approaches used for a broad spectrum of NLP tasks
requiring natural language understanding, generation, and reasoning.
Furthermore, we discuss the existing resources available for hybrid approaches
for NLP along with the challenges and future directions, offering a roadmap for
future research avenues. | cs.CL | Synergizing Machine Learning & Symbolic Methods: A Survey on Hybrid Approaches to Natural Language Processing |
2024-01-22T14:17:03Z | http://arxiv.org/pdf/2401.11969v3 | http://arxiv.org/abs/2401.11969v3 | Rrubaa Panchendrarajan, Arkaitz Zubiaga | 2024-03-18T16:49:59Z | cs.CL | Automated fact-checking has drawn considerable attention over the past few
decades due to the increase in the diffusion of misinformation on online
platforms. This is often carried out as a sequence of tasks comprising (i) the
detection of sentences circulating in online platforms which constitute claims
needing verification, followed by (ii) the verification process of those
claims. This survey focuses on the former, by discussing existing efforts
towards detecting claims needing fact-checking, with a particular focus on
multilingual data and methods. This is a challenging and fertile direction
where existing methods are yet far from matching human performance due to the
profoundly challenging nature of the issue. Especially, the dissemination of
information across multiple social platforms, articulated in multiple languages
and modalities demands more generalized solutions for combating misinformation.
Focusing on multilingual misinformation, we present a comprehensive survey of
existing multilingual claim detection research. We present state-of-the-art
multilingual claim detection research categorized into three key factors of the
problem, verifiability, priority, and similarity. Further, we present a
detailed overview of the existing multilingual datasets along with the
challenges and suggest possible future advancements. | cs.CL | Claim Detection for Automated Fact-checking: A Survey on Monolingual, Multilingual and Cross-Lingual Research |
2024-01-22T13:34:34Z | http://arxiv.org/pdf/2401.11944v2 | http://arxiv.org/abs/2401.11944v2 | Ge Zhang, Xinrun Du, Bei Chen, Yiming Liang, Tongxu Luo, Tianyu Zheng, Kang Zhu, Yuyang Cheng, Chunpu Xu, Shuyue Guo, Haoran Zhang, Xingwei Qu, Junjie Wang, Ruibin Yuan, Yizhi Li, Zekun Wang, Yudong Liu, Yu-Hsuan Tsai, Fengji Zhang, Chenghua Lin, Wenhao Huang, Wenhu Chen, Jie Fu | 2024-03-18T09:02:03Z | cs.CL, cs.AI, cs.CV | As the capabilities of large multimodal models (LMMs) continue to advance,
evaluating the performance of LMMs emerges as an increasing need. Additionally,
there is an even larger gap in evaluating the advanced knowledge and reasoning
abilities of LMMs in non-English contexts such as Chinese. We introduce CMMMU,
a new Chinese Massive Multi-discipline Multimodal Understanding benchmark
designed to evaluate LMMs on tasks demanding college-level subject knowledge
and deliberate reasoning in a Chinese context. CMMMU is inspired by and
strictly follows the annotation and analysis pattern of MMMU.
CMMMU includes 12k manually collected multimodal questions from college
exams, quizzes, and textbooks, covering six core disciplines: Art & Design,
Business, Science, Health & Medicine, Humanities & Social Science, and Tech &
Engineering, like its companion, MMMU. These questions span 30 subjects and
comprise 39 highly heterogeneous image types, such as charts, diagrams, maps,
tables, music sheets, and chemical structures.
CMMMU focuses on complex perception and reasoning with domain-specific
knowledge in the Chinese context. We evaluate 11 open-source LLMs and one
proprietary GPT-4V(ision). Even GPT-4V only achieves accuracies of 42%,
indicating a large space for improvement. CMMMU will boost the community to
build the next-generation LMMs towards expert artificial intelligence and
promote the democratization of LMMs by providing diverse language contexts. | cs.CL | CMMMU: A Chinese Massive Multi-discipline Multimodal Understanding Benchmark |
2024-01-22T13:33:53Z | http://arxiv.org/pdf/2401.11943v1 | http://arxiv.org/abs/2401.11943v1 | Jiawei Zhang, Tianyu Pang, Chao Du, Yi Ren, Bo Li, Min Lin | 2024-01-22T13:33:53Z | cs.LG, cs.CL, cs.CR, cs.CV, cs.MM | This technical report aims to fill a deficiency in the assessment of large
multimodal models (LMMs) by specifically examining the self-consistency of
their outputs when subjected to common corruptions. We investigate the
cross-modal interactions between text, image, and speech, encompassing four
essential generation tasks: text-to-image, image-to-text, text-to-speech, and
speech-to-text. We create a comprehensive benchmark, named MMCBench, that
covers more than 100 popular LMMs (totally over 150 model checkpoints). A
thorough evaluation under common corruptions is critical for practical
deployment and facilitates a better understanding of the reliability of
cutting-edge LMMs. The benchmarking code is available at
https://github.com/sail-sg/MMCBench | cs.LG | Benchmarking Large Multimodal Models against Common Corruptions |
2024-01-22T12:54:04Z | http://arxiv.org/pdf/2401.11911v6 | http://arxiv.org/abs/2401.11911v6 | Hexiang Tan, Fei Sun, Wanli Yang, Yuanzhuo Wang, Qi Cao, Xueqi Cheng | 2024-06-11T02:52:58Z | cs.CL, cs.AI | While auxiliary information has become a key to enhancing Large Language
Models (LLMs), relatively little is known about how LLMs merge these contexts,
specifically contexts generated by LLMs and those retrieved from external
sources. To investigate this, we formulate a systematic framework to identify
whether LLMs' responses are attributed to either generated or retrieved
contexts. To easily trace the origin of the response, we construct datasets
with conflicting contexts, i.e., each question is paired with both generated
and retrieved contexts, yet only one of them contains the correct answer. Our
experiments reveal a significant bias in several LLMs (GPT-4/3.5 and Llama2) to
favor generated contexts, even when they provide incorrect information. We
further identify two key factors contributing to this bias: i) contexts
generated by LLMs typically show greater similarity to the questions,
increasing their likelihood of being selected; ii) the segmentation process
used in retrieved contexts disrupts their completeness, thereby hindering their
full utilization in LLMs. Our analysis enhances the understanding of how LLMs
merge diverse contexts, offers valuable insights for advancing current LLM
augmentation methods, and highlights the risk of generated misinformation for
retrieval-augmented LLMs. | cs.CL | Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts? |
2024-01-22T12:11:55Z | http://arxiv.org/pdf/2401.11880v2 | http://arxiv.org/abs/2401.11880v2 | Zaibin Zhang, Yongting Zhang, Lijun Li, Hongzhi Gao, Lijun Wang, Huchuan Lu, Feng Zhao, Yu Qiao, Jing Shao | 2024-02-18T02:36:39Z | cs.CL, cs.AI, cs.CR, cs.MA | Multi-agent systems, when enhanced with Large Language Models (LLMs), exhibit
profound capabilities in collective intelligence. However, the potential misuse
of this intelligence for malicious purposes presents significant risks. To
date, comprehensive research on the safety issues associated with multi-agent
systems remains limited. In this paper, we explore these concerns through the
innovative lens of agent psychology, revealing that the dark psychological
states of agents constitute a significant threat to safety. To tackle these
concerns, we propose a comprehensive framework (PsySafe) grounded in agent
psychology, focusing on three key areas: firstly, identifying how dark
personality traits in agents can lead to risky behaviors; secondly, evaluating
the safety of multi-agent systems from the psychological and behavioral
perspectives, and thirdly, devising effective strategies to mitigate these
risks. Our experiments reveal several intriguing phenomena, such as the
collective dangerous behaviors among agents, agents' self-reflection when
engaging in dangerous behavior, and the correlation between agents'
psychological assessments and dangerous behaviors. We anticipate that our
framework and observations will provide valuable insights for further research
into the safety of multi-agent systems. We will make our data and code publicly
accessible at https://github.com/AI4Good24/PsySafe. | cs.CL | PsySafe: A Comprehensive Framework for Psychological-based Attack, Defense, and Evaluation of Multi-agent System Safety |
2024-01-22T11:37:18Z | http://arxiv.org/pdf/2401.11864v4 | http://arxiv.org/abs/2401.11864v4 | Xunyu Zhu, Jian Li, Yong Liu, Can Ma, Weiping Wang | 2024-02-01T18:16:04Z | cs.CL, cs.AI | This work addresses the challenge of democratizing advanced Large Language
Models (LLMs) by compressing their mathematical reasoning capabilities into
sub-billion parameter Small Language Models (SLMs) without compromising
performance. We introduce Equation-of-Thought Distillation (EoTD), a novel
technique that encapsulates the reasoning process into equation-based
representations to construct an EoTD dataset for fine-tuning SLMs.
Additionally, we propose the Ensemble Thoughts Distillation (ETD) framework to
enhance the reasoning performance of SLMs. This involves creating a reasoning
dataset with multiple thought processes, including Chain-of-Thought (CoT),
Program-of-Thought (PoT), and Equation-of-Thought (EoT), and using it for
fine-tuning. Our experimental findings demonstrate that EoTD significantly
boosts the reasoning abilities of SLMs, while ETD enables these models to
achieve state-of-the-art reasoning performance. | cs.CL | Distilling Mathematical Reasoning Capabilities into Small Language Models |
2024-01-22T11:15:07Z | http://arxiv.org/pdf/2401.11852v1 | http://arxiv.org/abs/2401.11852v1 | Martina Forster, Claudia Schulz, Prudhvi Nokku, Melicaalsadat Mirsafian, Jaykumar Kasundra, Stavroula Skylaki | 2024-01-22T11:15:07Z | cs.CL, cs.AI | Multi-Label Classification (MLC) is a common task in the legal domain, where
more than one label may be assigned to a legal document. A wide range of
methods can be applied, ranging from traditional ML approaches to the latest
Transformer-based architectures. In this work, we perform an evaluation of
different MLC methods using two public legal datasets, POSTURE50K and
EURLEX57K. By varying the amount of training data and the number of labels, we
explore the comparative advantage offered by different approaches in relation
to the dataset properties. Our findings highlight DistilRoBERTa and LegalBERT
as performing consistently well in legal MLC with reasonable computational
demands. T5 also demonstrates comparable performance while offering advantages
as a generative model in the presence of changing label sets. Finally, we show
that the CrossEncoder exhibits potential for notable macro-F1 score
improvements, albeit with increased computational costs. | cs.CL | The Right Model for the Job: An Evaluation of Legal Multi-Label Classification Baselines |
2024-01-22T10:57:09Z | http://arxiv.org/pdf/2401.11839v1 | http://arxiv.org/abs/2401.11839v1 | Ruoxi Xu, Yingfei Sun, Mengjie Ren, Shiguang Guo, Ruotong Pan, Hongyu Lin, Le Sun, Xianpei Han | 2024-01-22T10:57:09Z | cs.CL, cs.CY | Recent advancements in artificial intelligence, particularly with the
emergence of large language models (LLMs), have sparked a rethinking of
artificial general intelligence possibilities. The increasing human-like
capabilities of AI are also attracting attention in social science research,
leading to various studies exploring the combination of these two fields. In
this survey, we systematically categorize previous explorations in the
combination of AI and social science into two directions that share common
technical approaches but differ in their research objectives. The first
direction is focused on AI for social science, where AI is utilized as a
powerful tool to enhance various stages of social science research. While the
second direction is the social science of AI, which examines AI agents as
social entities with their human-like cognitive and linguistic capabilities. By
conducting a thorough review, particularly on the substantial progress
facilitated by recent advancements in large language models, this paper
introduces a fresh perspective to reassess the relationship between AI and
social science, provides a cohesive framework that allows researchers to
understand the distinctions and connections between AI for social science and
social science of AI, and also summarized state-of-art experiment simulation
platforms to facilitate research in these two directions. We believe that as AI
technology continues to advance and intelligent agents find increasing
applications in our daily lives, the significance of the combination of AI and
social science will become even more prominent. | cs.CL | AI for social science and social science of AI: A Survey |
2024-01-22T10:30:11Z | http://arxiv.org/pdf/2401.11819v2 | http://arxiv.org/abs/2401.11819v2 | Liang Xu, Hang Xue, Lei Zhu, Kangkang Zhao | 2024-02-02T02:35:13Z | cs.CL, cs.AI | We introduce SuperCLUE-Math6(SC-Math6), a new benchmark dataset to evaluate
the mathematical reasoning abilities of Chinese language models. SC-Math6 is
designed as an upgraded Chinese version of the GSM8K dataset with enhanced
difficulty, diversity, and application scope. It consists of over 2000
mathematical word problems requiring multi-step reasoning and providing natural
language solutions. We propose an innovative scheme to quantify the reasoning
capability of large models based on performance over problems with different
reasoning steps. Experiments on 13 representative Chinese models demonstrate a
clear stratification of reasoning levels, with top models like GPT-4 showing
superior performance. SC-Math6 fills the gap in Chinese mathematical reasoning
benchmarks and provides a comprehensive testbed to advance the intelligence of
Chinese language models. | cs.CL | SuperCLUE-Math6: Graded Multi-Step Math Reasoning Benchmark for LLMs in Chinese |
2024-01-22T10:26:14Z | http://arxiv.org/pdf/2401.11817v1 | http://arxiv.org/abs/2401.11817v1 | Ziwei Xu, Sanjay Jain, Mohan Kankanhalli | 2024-01-22T10:26:14Z | cs.CL, cs.AI, cs.LG | Hallucination has been widely recognized to be a significant drawback for
large language models (LLMs). There have been many works that attempt to reduce
the extent of hallucination. These efforts have mostly been empirical so far,
which cannot answer the fundamental question whether it can be completely
eliminated. In this paper, we formalize the problem and show that it is
impossible to eliminate hallucination in LLMs. Specifically, we define a formal
world where hallucination is defined as inconsistencies between a computable
LLM and a computable ground truth function. By employing results from learning
theory, we show that LLMs cannot learn all of the computable functions and will
therefore always hallucinate. Since the formal world is a part of the real
world which is much more complicated, hallucinations are also inevitable for
real world LLMs. Furthermore, for real world LLMs constrained by provable time
complexity, we describe the hallucination-prone tasks and empirically validate
our claims. Finally, using the formal world framework, we discuss the possible
mechanisms and efficacies of existing hallucination mitigators as well as the
practical implications on the safe deployment of LLMs. | cs.CL | Hallucination is Inevitable: An Innate Limitation of Large Language Models |
2024-01-22T09:41:05Z | http://arxiv.org/pdf/2401.11791v2 | http://arxiv.org/abs/2401.11791v2 | Ci-Siang Lin, Chien-Yi Wang, Yu-Chiang Frank Wang, Min-Hung Chen | 2024-03-11T04:01:50Z | cs.CV, cs.CL, cs.LG | Weakly-Supervised Semantic Segmentation (WSSS) aims to train segmentation
models using image data with only image-level supervision. Since precise
pixel-level annotations are not accessible, existing methods typically focus on
producing pseudo masks for training segmentation models by refining CAM-like
heatmaps. However, the produced heatmaps may capture only the discriminative
image regions of object categories or the associated co-occurring backgrounds.
To address the issues, we propose a Semantic Prompt Learning for WSSS (SemPLeS)
framework, which learns to effectively prompt the CLIP latent space to enhance
the semantic alignment between the segmented regions and the target object
categories. More specifically, we propose Contrastive Prompt Learning and
Prompt-guided Semantic Refinement to learn the prompts that adequately describe
and suppress the co-occurring backgrounds associated with each target object
category. In this way, SemPLeS can perform better semantic alignment between
object regions and the associated class labels, resulting in desired pseudo
masks for training the segmentation model. The proposed SemPLeS framework
achieves SOTA performance on the standard WSSS benchmarks, PASCAL VOC and MS
COCO, and shows compatibility with other WSSS methods. The source codes are
provided in the supplementary. | cs.CV | SemPLeS: Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation |
2024-01-22T08:47:50Z | http://arxiv.org/pdf/2402.01685v2 | http://arxiv.org/abs/2402.01685v2 | Yu Zhang, Mei Di, Haozheng Luo, Chenwei Xu, Richard Tzong-Han Tsai | 2024-02-06T06:03:13Z | cs.CL, cs.AI, cs.DB | We introduce SMUTF, a unique approach for large-scale tabular data schema
matching (SM), which assumes that supervised learning does not affect
performance in open-domain tasks, thereby enabling effective cross-domain
matching. This system uniquely combines rule-based feature engineering,
pre-trained language models, and generative large language models. In an
innovative adaptation inspired by the Humanitarian Exchange Language, we deploy
'generative tags' for each data column, enhancing the effectiveness of SM.
SMUTF exhibits extensive versatility, working seamlessly with any pre-existing
pre-trained embeddings, classification methods, and generative models.
Recognizing the lack of extensive, publicly available datasets for SM, we
have created and open-sourced the HDXSM dataset from the public humanitarian
data. We believe this to be the most exhaustive SM dataset currently available.
In evaluations across various public datasets and the novel HDXSM dataset,
SMUTF demonstrated exceptional performance, surpassing existing
state-of-the-art models in terms of accuracy and efficiency, and} improving the
F1 score by 11.84% and the AUC of ROC by 5.08%. | cs.CL | SMUTF: Schema Matching Using Generative Tags and Hybrid Features |
2024-01-22T07:58:31Z | http://arxiv.org/pdf/2402.01684v1 | http://arxiv.org/abs/2402.01684v1 | Chao Song, Zhihao Ye, Qiqiang Lin, Qiuying Peng, Jun Wang | 2024-01-22T07:58:31Z | cs.CL, cs.AI, cs.LG | With the productive evolution of large language models (LLMs) in the field of
natural language processing (NLP), tons of effort has been made to effectively
fine-tune common pre-trained LLMs to fulfill a variety of tasks in one or
multiple specific domain. In practice, there are two prevailing ways, in which
the adaptation can be achieved: (i) Multiple Independent Models: Pre-trained
LLMs are fine-tuned a few times independently using the corresponding training
samples from each task. (ii) An Integrated Model: Samples from all tasks are
employed to fine-tune a pre-trianed LLM unitedly. To address the high computing
cost and seesawing issue simultaneously, we propose a unified framework that
implements a 1 + N mutli-task fine-tuning pattern in LLMs using a novel
Customized Gate Control (CGC) Low-rank Adaptation (LoRA) algorithm. Our work
aims to take an advantage of both MTL (i.e., CGC) and PEFT (i.e., LoRA) scheme.
For a given cluster of tasks, we design an innovative layer that contains two
types of experts as additional trainable parameters to make LoRA be compatible
with MTL. To comprehensively evaluate the proposed framework, we conduct
well-designed experiments on two public datasets. The experimental results
demonstrate that the unified framework with CGC-LoRA modules achieves higher
evaluation scores than all benchmarks on both two datasets. | cs.CL | A Framework to Implement 1+N Multi-task Fine-tuning Pattern in LLMs Using the CGC-LORA Algorithm |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.