title stringlengths 23 147 | paper_url stringlengths 43 46 | authors listlengths 1 61 | abstract large_stringlengths 407 2.03k ⌀ | anthology_id stringlengths 17 20 | doi stringlengths 29 32 | award stringclasses 5
values | arxiv_id stringlengths 10 10 ⌀ | arxiv_id_source stringclasses 2
values | embedding listlengths 768 768 |
|---|---|---|---|---|---|---|---|---|---|
UniGen: Universal Domain Generalization for Sentiment Classification via Zero-shot Dataset Generation | https://aclanthology.org/2024.emnlp-main.1/ | [
"Juhwan Choi",
"Yeonghwa Kim",
"Seunguk Yu",
"JungMin Yun",
"YoungBin Kim"
] | Although pre-trained language models have exhibited great flexibility and versatility with prompt-based few-shot learning, they suffer from the extensive parameter size and limited applicability for inference. Recent studies have suggested that PLMs be used as dataset generators and a tiny task-specific model be traine... | 2024.emnlp-main.1 | 10.18653/v1/2024.emnlp-main.1 | null | 2405.01022 | title_snapshot | [
0.0017149518243968487,
-0.02875269204378128,
0.005253757815808058,
0.05100090801715851,
0.04158103093504906,
0.033116523176431656,
0.027755554765462875,
-0.0034482027404010296,
-0.004891787189990282,
0.009993375279009342,
-0.015148676000535488,
0.017401624470949173,
-0.08125966787338257,
-... |
Multi-News+: Cost-efficient Dataset Cleansing via LLM-based Data Annotation | https://aclanthology.org/2024.emnlp-main.2/ | [
"Juhwan Choi",
"JungMin Yun",
"Kyohoon Jin",
"YoungBin Kim"
] | The quality of the dataset is crucial for ensuring optimal performance and reliability of downstream task models. However, datasets often contain noisy data inadvertently included during the construction process. Numerous attempts have been made to correct this issue through human annotators. However, hiring and managi... | 2024.emnlp-main.2 | 10.18653/v1/2024.emnlp-main.2 | null | 2404.09682 | title_snapshot | [
-0.0020095123909413815,
-0.033672090619802475,
-0.013913259841501713,
0.02864348702132702,
0.06451702862977982,
-0.006991752423346043,
-0.0018460960127413273,
-0.0008937239181250334,
-0.0229488518089056,
-0.019842782989144325,
-0.0545404888689518,
0.05024814233183861,
-0.07741563767194748,
... |
FIZZ: Factual Inconsistency Detection by Zoom-in Summary and Zoom-out Document | https://aclanthology.org/2024.emnlp-main.3/ | [
"Joonho Yang",
"Seunghyun Yoon",
"ByeongJeong Kim",
"Hwanhee Lee"
] | Through the advent of pre-trained language models, there have been notable advancements in abstractive summarization systems. Simultaneously, a considerable number of novel methods for evaluating factual consistency in abstractive summarization systems has been developed. But these evaluation approaches incorporate sub... | 2024.emnlp-main.3 | 10.18653/v1/2024.emnlp-main.3 | null | 2404.11184 | title_snapshot | [
-0.01122517604380846,
-0.039762530475854874,
-0.022700412198901176,
0.04358302801847458,
0.057201121002435684,
-0.008018110878765583,
0.013553859665989876,
0.04053255915641785,
-0.015165826305747032,
-0.02630423568189144,
-0.055168163031339645,
0.03450951352715492,
-0.06177100911736488,
0.... |
Prompts have evil twins | https://aclanthology.org/2024.emnlp-main.4/ | [
"Rimon Melamed",
"Lucas Hurley McCabe",
"Tanay Wakhare",
"Yejin Kim",
"H. Howie Huang",
"Enric Boix-Adserà"
] | We discover that many natural-language prompts can be replaced by corresponding prompts that are unintelligible to humans but that provably elicit similar behavior in language models. We call these prompts “evil twins” because they are obfuscated and uninterpretable (evil), but at the same time mimic the functionality ... | 2024.emnlp-main.4 | 10.18653/v1/2024.emnlp-main.4 | null | 2311.07064 | title_snapshot | [
-0.011865461245179176,
-0.0227650236338377,
-0.004380557686090469,
0.043121688067913055,
0.026045093312859535,
0.02837158367037773,
0.048404842615127563,
0.004353696946054697,
-0.023975862190127373,
-0.008822941221296787,
-0.0581451915204525,
0.05037057772278786,
-0.06679778546094894,
-0.0... |
Table Question Answering for Low-resourced Indic Languages | https://aclanthology.org/2024.emnlp-main.5/ | [
"Vaishali Pal",
"Evangelos Kanoulas",
"Andrew Yates",
"Maarten de Rijke"
] | TableQA is the task of answering questions over tables of structured information, returning individual cells or tables as output. TableQA research has focused primarily on high-resource languages, leaving medium- and low-resource languages with little progress due to scarcity of annotated data and neural models. We add... | 2024.emnlp-main.5 | 10.18653/v1/2024.emnlp-main.5 | null | 2410.03576 | title_snapshot | [
-0.014048951677978039,
-0.00670273695141077,
-0.04015430435538292,
0.05095459893345833,
0.048122577369213104,
0.03539550304412842,
-0.0070173125714063644,
0.03294537588953972,
-0.03924081474542618,
0.025885215029120445,
-0.030849460512399673,
0.026802433654665947,
-0.07683910429477692,
-0.... |
ImageInWords: Unlocking Hyper-Detailed Image Descriptions | https://aclanthology.org/2024.emnlp-main.6/ | [
"Roopal Garg",
"Andrea Burns",
"Burcu Karagol Ayan",
"Yonatan Bitton",
"Ceslee Montgomery",
"Yasumasa Onoe",
"Andrew Bunner",
"Ranjay Krishna",
"Jason Michael Baldridge",
"Radu Soricut"
] | Despite the longstanding adage ”an image is worth a thousand words,” generating accurate hyper-detailed image descriptions remains unsolved. Trained on short web-scraped image-text, vision-language models often generate incomplete descriptions with visual inconsistencies. We address this via a novel data-centric approa... | 2024.emnlp-main.6 | 10.18653/v1/2024.emnlp-main.6 | null | 2405.02793 | title_snapshot | [
0.012172100134193897,
-0.017592933028936386,
0.025465264916419983,
0.06611450016498566,
0.01659165322780609,
0.002593592507764697,
0.02223847433924675,
0.008674210868775845,
-0.005494927056133747,
-0.020960332825779915,
-0.05100107565522194,
0.022211670875549316,
-0.0521635003387928,
-0.00... |
LLM-Based Agent Society Investigation: Collaboration and Confrontation in Avalon Gameplay | https://aclanthology.org/2024.emnlp-main.7/ | [
"Yihuai Lan",
"Zhiqiang Hu",
"Lei Wang",
"Yang Wang",
"Deheng Ye",
"Peilin Zhao",
"Ee-Peng Lim",
"Hui Xiong",
"Hao Wang"
] | This paper explores the open research problem of understanding the social behaviors of LLM-based agents. Using Avalon as a testbed, we employ system prompts to guide LLM agents in gameplay. While previous studies have touched on gameplay with LLM agents, research on their social behaviors is lacking. We propose a novel... | 2024.emnlp-main.7 | 10.18653/v1/2024.emnlp-main.7 | null | 2310.14985 | title_snapshot | [
-0.0333283431828022,
-0.036207471042871475,
0.02139788307249546,
0.0073481048457324505,
0.036396324634552,
-0.014940747059881687,
0.028011633083224297,
0.033349666744470596,
-0.015190323814749718,
-0.03353190794587135,
-0.04320499673485756,
0.05300271511077881,
-0.04404548928141594,
0.0001... |
When LLMs Meets Acoustic Landmarks: An Efficient Approach to Integrate Speech into Large Language Models for Depression Detection | https://aclanthology.org/2024.emnlp-main.8/ | [
"Xiangyu Zhang",
"Hexin Liu",
"Kaishuai Xu",
"Qiquan Zhang",
"Daijiao Liu",
"Beena Ahmed",
"Julien Epps"
] | Depression is a critical concern in global mental health, prompting extensive research into AI-based detection methods. Among various AI technologies, Large Language Models (LLMs) stand out for their versatility in healthcare applications. However, the application of LLMs in the identification and analysis of depressiv... | 2024.emnlp-main.8 | 10.18653/v1/2024.emnlp-main.8 | null | 2402.13276 | title_snapshot | [
-0.021929575130343437,
0.014992046169936657,
0.008902452886104584,
-0.016330374404788017,
0.054388027638196945,
0.05404583737254143,
0.04977297782897949,
0.02997325360774994,
-0.029019266366958618,
-0.04007256031036377,
-0.03750837221741676,
0.029263382777571678,
-0.04333174228668213,
0.01... |
Speaking in Wavelet Domain: A Simple and Efficient Approach to Speed up Speech Diffusion Model | https://aclanthology.org/2024.emnlp-main.9/ | [
"Xiangyu Zhang",
"Daijiao Liu",
"Hexin Liu",
"Qiquan Zhang",
"Hanyu Meng",
"Leibny Paola Garcia",
"Eng Siong Chng",
"Lina Yao"
] | Recently, Denoising Diffusion Probabilistic Models (DDPMs) have attained leading performances across a diverse range of generative tasks. However, in the field of speech synthesis, although DDPMs exhibit impressive performance, their prolonged training duration and substantial inference costs hinder practical deploymen... | 2024.emnlp-main.9 | 10.18653/v1/2024.emnlp-main.9 | null | 2402.10642 | title_snapshot | [
-0.021708736196160316,
-0.009302044287323952,
-0.01628449559211731,
0.05679318681359291,
0.049281127750873566,
0.049517352133989334,
0.05606505274772644,
-0.009940966963768005,
-0.01692035235464573,
-0.07105976343154907,
-0.007196195423603058,
0.004692062735557556,
-0.022689098492264748,
0... |
Hateful Word in Context Classification | https://aclanthology.org/2024.emnlp-main.10/ | [
"Sanne Hoeken",
"Sina Zarrieß",
"Özge Alacam"
] | Hate speech detection is a prevalent research field, yet it remains underexplored at the level of word meaning. This is significant, as terms used to convey hate often involve non-standard or novel usages which might be overlooked by commonly leveraged LMs trained on general language use. In this paper, we introduce th... | 2024.emnlp-main.10 | 10.18653/v1/2024.emnlp-main.10 | null | null | null | [
0.009924206882715225,
-0.009501592256128788,
0.00035956897772848606,
0.006984382402151823,
0.02530338615179062,
0.035933785140514374,
0.0702664852142334,
0.02054218016564846,
0.01779152825474739,
-0.01046033576130867,
-0.026375409215688705,
0.03141966462135315,
-0.05394577234983444,
-0.022... |
Eyes Don’t Lie: Subjective Hate Annotation and Detection with Gaze | https://aclanthology.org/2024.emnlp-main.11/ | [
"Özge Alacam",
"Sanne Hoeken",
"Sina Zarrieß"
] | Hate speech is a complex and subjective phenomenon. In this paper, we present a dataset (GAZE4HATE) that provides gaze data collected in a hate speech annotation experiment. We study whether the gaze of an annotator provides predictors of their subjective hatefulness rating, and how gaze features can improve Hate Speec... | 2024.emnlp-main.11 | 10.18653/v1/2024.emnlp-main.11 | null | null | null | [
0.01897553913295269,
0.01821787655353546,
0.0008059365791268647,
0.001613155473023653,
0.00350076868198812,
0.028132550418376923,
0.05954068526625633,
0.029538657516241074,
0.011226420290768147,
-0.012639285065233707,
-0.03792141005396843,
0.025854753330349922,
-0.06656936556100845,
-0.031... |
NumeroLogic: Number Encoding for Enhanced LLMs’ Numerical Reasoning | https://aclanthology.org/2024.emnlp-main.12/ | [
"Eli Schwartz",
"Leshem Choshen",
"Joseph Shtok",
"Sivan Doveh",
"Leonid Karlinsky",
"Assaf Arbelle"
] | Language models struggle with handling numerical data and performing arithmetic operations. We hypothesize that this limitation can be partially attributed to non-intuitive textual numbers representation. When a digit is read or generated by a causal language model it does not know its place value (e.g. thousands vs. h... | 2024.emnlp-main.12 | 10.18653/v1/2024.emnlp-main.12 | null | 2404.00459 | title_snapshot | [
-0.0276812557131052,
-0.01044443529099226,
-0.03333905711770058,
0.027377773076295853,
0.05925514176487923,
0.034941621124744415,
0.033393070101737976,
0.018861450254917145,
-0.07312265783548355,
-0.007819403894245625,
-0.006173080299049616,
0.04422536864876747,
-0.044974833726882935,
0.00... |
“Thinking” Fair and Slow: On the Efficacy of Structured Prompts for Debiasing Language Models | https://aclanthology.org/2024.emnlp-main.13/ | [
"Shaz Furniturewala",
"Surgan Jandial",
"Abhinav Java",
"Pragyan Banerjee",
"Simra Shahid",
"Sumit Bhatia",
"Kokil Jaidka"
] | Existing debiasing techniques are typically training-based or require access to the model’s internals and output distributions, so they are inaccessible to end-users looking to adapt LLM outputs for their particular needs. In this study, we examine whether structured prompting techniques can offer opportunities for fai... | 2024.emnlp-main.13 | 10.18653/v1/2024.emnlp-main.13 | null | 2405.10431 | title_snapshot | [
-0.000194534586626105,
-0.03318984434008598,
-0.029360655695199966,
0.036905355751514435,
0.030691223219037056,
0.00014148953778203577,
0.0026290903333574533,
0.021333638578653336,
-0.018236642703413963,
0.006844868417829275,
-0.005346592515707016,
0.04107525572180748,
-0.07163777947425842,
... |
A Usage-centric Take on Intent Understanding in E-Commerce | https://aclanthology.org/2024.emnlp-main.14/ | [
"Wendi Zhou",
"Tianyi Li",
"Pavlos Vougiouklis",
"Mark Steedman",
"Jeff Z. Pan"
] | Identifying and understanding user intents is a pivotal task for E-Commerce. Despite its essential role in product recommendation and business user profiling analysis, intent understanding has not been consistently defined or accurately benchmarked. In this paper, we focus on predicative user intents as “how a customer... | 2024.emnlp-main.14 | 10.18653/v1/2024.emnlp-main.14 | null | 2402.14901 | title_snapshot | [
0.010724506340920925,
-0.02547212317585945,
-0.016381140798330307,
0.011719685979187489,
0.04866178333759308,
-0.004925227258354425,
0.04276296868920326,
0.02097989432513714,
0.03967941179871559,
-0.0015979567542672157,
-0.03890187293291092,
0.03354218974709511,
-0.051458317786455154,
-0.0... |
Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs | https://aclanthology.org/2024.emnlp-main.15/ | [
"Oded Ovadia",
"Menachem Brief",
"Moshik Mishaeli",
"Oren Elisha"
] | Large language models (LLMs) encapsulate a vast amount of factual information within their pre-trained weights, as evidenced by their ability to answer diverse questions across different domains. However, this knowledge is inherently limited, relying heavily on the characteristics of the training data. Consequently, us... | 2024.emnlp-main.15 | 10.18653/v1/2024.emnlp-main.15 | null | 2312.05934 | title_snapshot | [
-0.012847757898271084,
-0.03172774985432625,
-0.003134992439299822,
0.04050429165363312,
0.06851837038993835,
-0.020536018535494804,
0.03030608594417572,
0.007690025959163904,
-0.026438331231474876,
0.009751797653734684,
-0.04725923761725426,
0.05725478753447533,
-0.05057966709136963,
0.00... |
Systematic Biases in LLM Simulations of Debates | https://aclanthology.org/2024.emnlp-main.16/ | [
"Amir Taubenfeld",
"Yaniv Dover",
"Roi Reichart",
"Ariel Goldstein"
] | The emergence of Large Language Models (LLMs), has opened exciting possibilities for constructing computational simulations designed to replicate human behavior accurately. Current research suggests that LLM-based agents become increasingly human-like in their performance, sparking interest in using these AI agents as ... | 2024.emnlp-main.16 | 10.18653/v1/2024.emnlp-main.16 | null | 2402.04049 | title_snapshot | [
-0.018240399658679962,
-0.0016180460806936026,
-0.01900377683341503,
0.018337754532694817,
0.039099469780921936,
-0.004925011657178402,
0.03391602635383606,
0.029522713273763657,
-0.026166191324591637,
-0.015301372855901718,
-0.015936248004436493,
0.02688721939921379,
-0.06102721393108368,
... |
Studying and Mitigating Biases in Sign Language Understanding Models | https://aclanthology.org/2024.emnlp-main.17/ | [
"Katherine Atwell",
"Danielle Bragg",
"Malihe Alikhani"
] | Ensuring that the benefits of sign language technologies are distributed equitably among all community members is crucial. Thus, it is important to address potential biases and inequities that may arise from the design or use of these resources. Crowd-sourced sign language datasets, such as the ASL Citizen dataset, are... | 2024.emnlp-main.17 | 10.18653/v1/2024.emnlp-main.17 | null | 2410.05206 | title_snapshot | [
0.010783527977764606,
-0.01660417951643467,
-0.033650852739810944,
0.016985107213258743,
0.0183374285697937,
0.02458139695227146,
0.055571772158145905,
0.027599256485700607,
-0.03454570472240448,
-0.03636588901281357,
-0.01952407695353031,
0.017391808331012726,
-0.09831400215625763,
-0.026... |
Uncertainty in Language Models: Assessment through Rank-Calibration | https://aclanthology.org/2024.emnlp-main.18/ | [
"Xinmeng Huang",
"Shuo Li",
"Mengxin Yu",
"Matteo Sesia",
"Hamed Hassani",
"Insup Lee",
"Osbert Bastani",
"Edgar Dobriban"
] | Language Models (LMs) have shown promising performance in natural language generation. However, as LMs often generate incorrect or hallucinated responses, it is crucial to correctly quantify their uncertainty in responding to given inputs. In addition to verbalized confidence elicited via prompting, many uncertainty me... | 2024.emnlp-main.18 | 10.18653/v1/2024.emnlp-main.18 | null | 2404.03163 | title_snapshot | [
-0.007512544747442007,
-0.00944901630282402,
-0.04242659732699394,
0.03628181666135788,
0.03090880624949932,
0.033431526273489,
0.0465862900018692,
0.019474228844046593,
-0.03366345167160034,
-0.018249226734042168,
-0.02043185755610466,
0.04982484132051468,
-0.04082227870821953,
-0.0211963... |
RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning | https://aclanthology.org/2024.emnlp-main.19/ | [
"Junjie Ye",
"Yilong Wu",
"Songyang Gao",
"Caishuang Huang",
"Sixian Li",
"Guanyu Li",
"Xiaoran Fan",
"Qi Zhang",
"Tao Gui",
"Xuanjing Huang"
] | Tool learning has generated widespread interest as a vital means of interaction between Large Language Models (LLMs) and the physical world. Current research predominantly emphasizes LLMs’ capacity to utilize tools in well-structured environments while overlooking their stability when confronted with the inevitable noi... | 2024.emnlp-main.19 | 10.18653/v1/2024.emnlp-main.19 | null | 2401.08326 | title_snapshot | [
-0.011804393492639065,
-0.01712522841989994,
-0.02232571318745613,
0.018381861969828606,
0.022865338250994682,
0.0396525114774704,
0.03882116824388504,
0.005540458485484123,
-0.027536604553461075,
-0.016956845298409462,
-0.0068681007251143456,
0.05428723990917206,
-0.05928239971399307,
-0.... |
Learning Planning-based Reasoning by Trajectories Collection and Process Reward Synthesizing | https://aclanthology.org/2024.emnlp-main.20/ | [
"Fangkai Jiao",
"Chengwei Qin",
"Zhengyuan Liu",
"Nancy F. Chen",
"Shafiq Joty"
] | Large Language Models (LLMs) have demonstrated significant potential in handling complex reasoning tasks through step-by-step rationale generation. However, recent studies have raised concerns regarding the hallucination and flaws in their reasoning process. Substantial efforts are being made to improve the reliability... | 2024.emnlp-main.20 | 10.18653/v1/2024.emnlp-main.20 | Outstanding Paper Award | 2402.00658 | title_snapshot | [
-0.037391431629657745,
-0.007834477350115776,
-0.009641198441386223,
0.017568597570061684,
0.06828830391168594,
0.0045142872259020805,
0.0036984533071517944,
0.016121279448270798,
-0.020529642701148987,
0.0017093488713726401,
-0.03386319801211357,
0.03256232663989067,
-0.042738549411296844,
... |
Scaling Properties of Speech Language Models | https://aclanthology.org/2024.emnlp-main.21/ | [
"Santiago Cuervo",
"Ricard Marxer"
] | Speech Language Models (SLMs) aim to learn language from raw audio, without textual resources. Despite significant advances, our current models exhibit weak syntax and semantic abilities. However, if the scaling properties of neural language models hold for the speech modality, these abilities will improve as the amoun... | 2024.emnlp-main.21 | 10.18653/v1/2024.emnlp-main.21 | null | 2404.00685 | title_snapshot | [
-0.050090208649635315,
-0.008101047948002815,
0.0005392683669924736,
0.0054703750647604465,
0.03469396382570267,
0.05474633350968361,
0.03620333597064018,
0.029180793091654778,
-0.04571075364947319,
0.007725514937192202,
-0.0024780507665127516,
0.022446749731898308,
-0.04369344562292099,
0... |
“We Demand Justice!”: Towards Social Context Grounding of Political Texts | https://aclanthology.org/2024.emnlp-main.22/ | [
"Rajkumar Pujari",
"Chengfei Wu",
"Dan Goldwasser"
] | Political discourse on social media often contains similar language with opposing intended meanings. For example, the phrase thoughts and prayers, is used to express sympathy for mass shooting victims, as well as satirically criticize the lack of legislative action on gun control. Understanding such discourse fully by ... | 2024.emnlp-main.22 | 10.18653/v1/2024.emnlp-main.22 | null | 2311.09106 | title_snapshot | [
0.016313351690769196,
-0.006166904699057341,
-0.022479377686977386,
0.050259239971637726,
0.0026673385873436928,
0.02023688517510891,
0.03796743229031563,
0.020626602694392204,
-0.005386559758335352,
-0.003991648089140654,
-0.022338004782795906,
0.01323174312710762,
-0.06759382784366608,
-... |
An Experimental Analysis on Evaluating Patent Citations | https://aclanthology.org/2024.emnlp-main.23/ | [
"Rabindra Nath Nandi",
"Suman Kalyan Maity",
"Brian Uzzi",
"Sourav Medya"
] | The patent citation count is a good indicator of patent quality. This often generates monetary value for the inventors and organizations. However, the factors that influence a patent receiving high citations over the year are still not well understood. With the patents over the past two decades, we study the problem of... | 2024.emnlp-main.23 | 10.18653/v1/2024.emnlp-main.23 | null | null | null | [
-0.017039671540260315,
-0.021904291585087776,
-0.009908129461109638,
0.03652031719684601,
0.06568015366792679,
0.0026943490374833345,
-0.017583463340997696,
0.0061464980244636536,
-0.0005973656079731882,
-0.026202991604804993,
0.0071999612264335155,
0.0018239787314087152,
-0.0543496757745742... |
Fine-Tuning Large Language Models to Translate: Will a Touch of Noisy Data in Misaligned Languages Suffice? | https://aclanthology.org/2024.emnlp-main.24/ | [
"Dawei Zhu",
"Pinzhen Chen",
"Miaoran Zhang",
"Barry Haddow",
"Xiaoyu Shen",
"Dietrich Klakow"
] | Traditionally, success in multilingual machine translation can be attributed to three key factors in training data: large volume, diverse translation directions, and high quality. In the current practice of fine-tuning large language models (LLMs) for translation, we revisit the importance of these factors. We find tha... | 2024.emnlp-main.24 | 10.18653/v1/2024.emnlp-main.24 | null | 2404.14122 | title_snapshot | [
-0.020577697083353996,
-0.02524765208363533,
-0.004550510551780462,
0.0027362871915102005,
0.06328049302101135,
0.02852994203567505,
0.019825270399451256,
0.03678857162594795,
-0.023917075246572495,
-0.0036453527864068747,
-0.02734196186065674,
0.0432608388364315,
-0.05967579036951065,
0.0... |
Consolidating Ranking and Relevance Predictions of Large Language Models through Post-Processing | https://aclanthology.org/2024.emnlp-main.25/ | [
"Le Yan",
"Zhen Qin",
"Honglei Zhuang",
"Rolf Jagerman",
"Xuanhui Wang",
"Michael Bendersky",
"Harrie Oosterhuis"
] | The powerful generative abilities of large language models (LLMs) show potential in generating relevance labels for search applications. Previous work has found that directly asking about relevancy, such as "*How relevant is document A to query Q?*”, results in suboptimal ranking. Instead, the pairwise-ranking promptin... | 2024.emnlp-main.25 | 10.18653/v1/2024.emnlp-main.25 | null | 2404.11791 | title_snapshot | [
-0.03402328863739967,
-0.02294231206178665,
-0.010715801268815994,
0.03048197738826275,
0.03288528695702553,
-0.0016929915873333812,
0.008676636964082718,
0.0230092816054821,
-0.023142585530877113,
0.006214255467057228,
-0.03707879036664963,
0.0355556383728981,
-0.03722186014056206,
-0.000... |
Strength Lies in Differences! Improving Strategy Planning for Non-collaborative Dialogues via Diversified User Simulation | https://aclanthology.org/2024.emnlp-main.26/ | [
"Tong Zhang",
"Chen Huang",
"Yang Deng",
"Hongru Liang",
"Jia Liu",
"Zujie Wen",
"Wenqiang Lei",
"Tat-Seng Chua"
] | We investigate non-collaborative dialogue agents, which are expected to engage in strategic conversations with diverse users, for securing a mutual agreement that leans favorably towards the system’s objectives. This poses two main challenges for existing dialogue agents: 1) The inability to integrate user-specific cha... | 2024.emnlp-main.26 | 10.18653/v1/2024.emnlp-main.26 | null | 2403.06769 | title_snapshot | [
-0.022178757935762405,
-0.047799140214920044,
-0.024299703538417816,
0.042464904487133026,
0.0369887538254261,
0.02245085872709751,
0.026435986161231995,
0.005526451859623194,
-0.02977527678012848,
-0.047041378915309906,
-0.034436166286468506,
0.0466453842818737,
-0.07345865666866302,
-0.0... |
Impeding LLM-assisted Cheating in Introductory Programming Assignments via Adversarial Perturbation | https://aclanthology.org/2024.emnlp-main.27/ | [
"Saiful Islam Salim",
"Rubin Yuchan Yang",
"Alexander Cooper",
"Suryashree Ray",
"Saumya Debray",
"Sazzadur Rahaman"
] | While Large language model (LLM)-based programming assistants such as CoPilot and ChatGPT can help improve the productivity of professional software developers, they can also facilitate cheating in introductory computer programming courses. Assuming instructors have limited control over the industrial-strength models, ... | 2024.emnlp-main.27 | 10.18653/v1/2024.emnlp-main.27 | null | 2410.09318 | title_snapshot | [
0.008234848268330097,
-0.023087510839104652,
-0.05035199597477913,
0.021868925541639328,
0.06356030702590942,
-0.001197829027660191,
0.039302632212638855,
0.023748567327857018,
-0.04271816462278366,
0.006534149870276451,
-0.04449529945850372,
0.03999568521976471,
-0.08427485078573227,
-0.0... |
Clustering and Ranking: Diversity-preserved Instruction Selection through Expert-aligned Quality Estimation | https://aclanthology.org/2024.emnlp-main.28/ | [
"Yuan Ge",
"Yilun Liu",
"Chi Hu",
"Weibin Meng",
"Shimin Tao",
"Xiaofeng Zhao",
"Mahong Xia",
"Zhang Li",
"Boxing Chen",
"Hao Yang",
"Bei Li",
"Tong Xiao",
"JingBo Zhu"
] | With contributions from the open-source community, a vast amount of instruction tuning (IT) data has emerged. Given the significant resource allocation required by training and evaluating models, it is advantageous to have an efficient method for selecting high-quality IT data. However, existing methods for instruction... | 2024.emnlp-main.28 | 10.18653/v1/2024.emnlp-main.28 | null | 2402.18191 | title_snapshot | [
-0.010746875777840614,
-0.031150080263614655,
-0.029200708493590355,
0.06467533111572266,
0.04803904891014099,
0.043795328587293625,
0.02659468911588192,
0.001828465610742569,
0.008710398338735104,
-0.03907756879925728,
0.006032293662428856,
0.03121066279709339,
-0.03536512330174446,
-0.00... |
On the Influence of Gender and Race in Romantic Relationship Prediction from Large Language Models | https://aclanthology.org/2024.emnlp-main.29/ | [
"Abhilasha Sancheti",
"Haozhe An",
"Rachel Rudinger"
] | We study the presence of heteronormative biases and prejudice against interracial romantic relationships in large language models by performing controlled name-replacement experiments for the task of relationship prediction. We show that models are less likely to predict romantic relationships for (a) same-gender chara... | 2024.emnlp-main.29 | 10.18653/v1/2024.emnlp-main.29 | null | 2410.03996 | title_snapshot | [
-0.04303723946213722,
0.01665876805782318,
-0.04097968339920044,
0.0424061082303524,
0.02973676100373268,
0.03141464293003082,
0.05734822526574135,
0.028944609686732292,
-0.003513642121106386,
-0.00019798935682047158,
-0.022225771099328995,
0.01999766379594803,
-0.08742625266313553,
-0.024... |
EmphAssess : a Prosodic Benchmark on Assessing Emphasis Transfer in Speech-to-Speech Models | https://aclanthology.org/2024.emnlp-main.30/ | [
"Maureen de Seyssel",
"Antony D’Avirro",
"Adina Williams",
"Emmanuel Dupoux"
] | We introduce EmphAssess, a prosodic benchmark designed to evaluate the capability of speech-to-speech models to encode and reproduce prosodic emphasis. We apply this to two tasks: speech resynthesis and speech-to-speech translation. In both cases, the benchmark evaluates the ability of the model to encode emphasis in t... | 2024.emnlp-main.30 | 10.18653/v1/2024.emnlp-main.30 | null | 2312.14069 | title_snapshot | [
0.0034651984460651875,
-0.011800878681242466,
-0.015373989008367062,
0.02747802622616291,
0.0021005854941904545,
0.05876759812235832,
0.04290169104933739,
0.03926766291260719,
-0.013435482047498226,
-0.03262118622660637,
-0.006828519515693188,
0.017012502998113632,
-0.03673870116472244,
-0... |
On Fake News Detection with LLM Enhanced Semantics Mining | https://aclanthology.org/2024.emnlp-main.31/ | [
"Xiaoxiao Ma",
"Yuchen Zhang",
"Kaize Ding",
"Jian Yang",
"Jia Wu",
"Hao Fan"
] | Large language models (LLMs) have emerged as valuable tools for enhancing textual features in various text-related tasks. Despite their superiority in capturing the lexical semantics between tokens for text analysis, our preliminary study on two popular LLMs, i.e., ChatGPT and Llama2, showcases that simply applying the... | 2024.emnlp-main.31 | 10.18653/v1/2024.emnlp-main.31 | null | null | null | [
0.019268086180090904,
-0.03570093959569931,
-0.006936622783541679,
0.04322608932852745,
0.05763084441423416,
-0.008632180280983448,
0.017947308719158173,
0.04279498755931854,
-0.009282417595386505,
0.025272564962506294,
-0.04495418444275856,
0.04169180989265442,
-0.06870732456445694,
0.017... |
On Sensitivity of Learning with Limited Labelled Data to the Effects of Randomness: Impact of Interactions and Systematic Choices | https://aclanthology.org/2024.emnlp-main.32/ | [
"Branislav Pecher",
"Ivan Srba",
"Maria Bielikova"
] | While learning with limited labelled data can effectively deal with a lack of labels, it is also sensitive to the effects of uncontrolled randomness introduced by so-called randomness factors (i.e., non-deterministic decisions such as choice or order of samples). We propose and formalise a method to systematically inve... | 2024.emnlp-main.32 | 10.18653/v1/2024.emnlp-main.32 | null | 2402.12817 | title_snapshot | [
-0.0174762811511755,
-0.01586693339049816,
-0.020565791055560112,
0.046795256435871124,
0.045805610716342926,
0.0029013953171670437,
0.02818242646753788,
-0.00974339060485363,
-0.02964191883802414,
-0.023243144154548645,
-0.020935604348778725,
0.02523079141974449,
-0.039239417761564255,
-0... |
Evaluating the Instruction-Following Robustness of Large Language Models to Prompt Injection | https://aclanthology.org/2024.emnlp-main.33/ | [
"Zekun Li",
"Baolin Peng",
"Pengcheng He",
"Xifeng Yan"
] | Large Language Models (LLMs) have demonstrated exceptional proficiency in instruction-following, making them increasingly integral to various applications. However, this capability introduces the risk of prompt injection attacks, where malicious instructions are embedded in the input to trigger unintended actions or co... | 2024.emnlp-main.33 | 10.18653/v1/2024.emnlp-main.33 | null | 2308.10819 | title_snapshot | [
-0.033727142959833145,
-0.015257620252668858,
-0.027674537152051926,
0.0389782153069973,
0.04211292415857315,
-0.008472678251564503,
0.0595538429915905,
0.026855794712901115,
-0.022045478224754333,
0.01387803629040718,
-0.022893061861395836,
0.025260871276259422,
-0.03863407298922539,
-0.0... |
A Study of Nationality Bias in Names and Perplexity using Off-the-Shelf Affect-related Tweet Classifiers | https://aclanthology.org/2024.emnlp-main.34/ | [
"Valentin Barriere",
"Sebastian Cifuentes"
] | In this paper, we apply a method to quantify biases associated with named entities from various countries. We create counterfactual examples with small perturbations on target-domain data instead of relying on templates or specific datasets for bias detection. On widely used classifiers for subjectivity analysis, inclu... | 2024.emnlp-main.34 | 10.18653/v1/2024.emnlp-main.34 | null | 2407.01834 | title_snapshot | [
0.029548708349466324,
0.00041251981747336686,
-0.021335158497095108,
0.004852635785937309,
-0.008561305701732635,
0.023679662495851517,
0.04610983654856682,
0.028015147894620895,
0.0006495363777503371,
-0.004775622859597206,
-0.01797761209309101,
0.015634603798389435,
-0.08329026401042938,
... |
Mitigating the Alignment Tax of RLHF | https://aclanthology.org/2024.emnlp-main.35/ | [
"Yong Lin",
"Hangyu Lin",
"Wei Xiong",
"Shizhe Diao",
"Jianmeng Liu",
"Jipeng Zhang",
"Rui Pan",
"Haoxiang Wang",
"Wenbin Hu",
"Hanning Zhang",
"Hanze Dong",
"Renjie Pi",
"Han Zhao",
"Nan Jiang",
"Heng Ji",
"Yuan Yao",
"Tong Zhang"
] | LLMs acquire a wide range of abilities during pre-training, but aligning LLMs under Reinforcement Learning with Human Feedback (RLHF) can lead to forgetting pretrained abilities, which is also known as the alignment tax. To investigate alignment tax, we conducted experiments with existing RLHF algorithms using OpenLLaM... | 2024.emnlp-main.35 | 10.18653/v1/2024.emnlp-main.35 | null | 2309.06256 | title_snapshot | [
0.007732356432825327,
-0.0007145439740270376,
0.006499634124338627,
0.03145529702305794,
0.024290001019835472,
0.004673075396567583,
0.03494206815958023,
-0.012306085787713528,
-0.030279645696282387,
-0.001975233433768153,
0.004460880532860756,
0.043258924037218094,
-0.09294570982456207,
-... |
Evaluating Readability and Faithfulness of Concept-based Explanations | https://aclanthology.org/2024.emnlp-main.36/ | [
"Meng Li",
"Haoran Jin",
"Ruixuan Huang",
"Zhihao Xu",
"Defu Lian",
"Zijia Lin",
"Di Zhang",
"Xiting Wang"
] | With the growing popularity of general-purpose Large Language Models (LLMs), comes a need for more global explanations of model behaviors. Concept-based explanations arise as a promising avenue for explaining high-level patterns learned by LLMs. Yet their evaluation poses unique challenges, especially due to their non-... | 2024.emnlp-main.36 | 10.18653/v1/2024.emnlp-main.36 | null | 2404.18533 | title_snapshot | [
-0.03806303068995476,
-0.013398982584476471,
-0.024615023285150528,
0.0608944445848465,
0.04312711954116821,
0.008220904506742954,
0.025572579354047775,
0.010434833355247974,
-0.0038080387748777866,
-0.004224848002195358,
-0.013821885921061039,
0.044926341623067856,
-0.03322701156139374,
0... |
Personality-aware Student Simulation for Conversational Intelligent Tutoring Systems | https://aclanthology.org/2024.emnlp-main.37/ | [
"Zhengyuan Liu",
"Stella Xin Yin",
"Geyu Lin",
"Nancy F. Chen"
] | Intelligent Tutoring Systems (ITSs) can provide personalized and self-paced learning experience. The emergence of large language models (LLMs) further enables better human-machine interaction, and facilitates the development of conversational ITSs in various disciplines such as math and language learning. In dialogic t... | 2024.emnlp-main.37 | 10.18653/v1/2024.emnlp-main.37 | null | 2404.06762 | title_snapshot | [
-0.01763940416276455,
-0.02896607480943203,
0.014523457735776901,
0.02984396368265152,
0.05969047173857689,
0.026828352361917496,
0.03124176897108555,
0.04326864704489708,
-0.02162780798971653,
-0.012513596564531326,
-0.03885137662291527,
0.0218006893992424,
-0.01856297440826893,
-0.006482... |
MSI-Agent: Incorporating Multi-Scale Insight into Embodied Agents for Superior Planning and Decision-Making | https://aclanthology.org/2024.emnlp-main.38/ | [
"Dayuan Fu",
"Biqing Qi",
"Yihuai Gao",
"Che Jiang",
"Guanting Dong",
"Bowen Zhou"
] | Insight gradually becomes a crucial form of long-term memory for an agent. However, the emergence of irrelevant insight and the lack of general insight can greatly undermine the effectiveness of insight. To solve this problem, in this paper, we introduce **M**ulti-**S**cale **I**nsight Agent (MSI-Agent), an embodied ag... | 2024.emnlp-main.38 | 10.18653/v1/2024.emnlp-main.38 | null | 2409.16686 | title_snapshot | [
-0.04675653204321861,
-0.02001873590052128,
0.007013967260718346,
0.013164267875254154,
0.04451807588338852,
0.025647131726145744,
0.021640701219439507,
-0.008912179619073868,
-0.03543721139431,
-0.038664527237415314,
-0.011601563543081284,
0.02889353223145008,
-0.052331969141960144,
-0.00... |
CoCoLoFa: A Dataset of News Comments with Common Logical Fallacies Written by LLM-Assisted Crowds | https://aclanthology.org/2024.emnlp-main.39/ | [
"Min-Hsuan Yeh",
"Ruyuan Wan",
"Ting-Hao Kenneth Huang"
] | Detecting logical fallacies in texts can help users spot argument flaws, but automating this detection is not easy. Manually annotating fallacies in large-scale, real-world text data to create datasets for developing and validating detection models is costly. This paper introduces CoCoLoFa, the largest known logical fa... | 2024.emnlp-main.39 | 10.18653/v1/2024.emnlp-main.39 | null | 2410.03457 | title_snapshot | [
0.018528062850236893,
-0.025087378919124603,
-0.01671721041202545,
0.055311575531959534,
0.04684498906135559,
0.01442316360771656,
-0.0000986283557722345,
0.03692125529050827,
-0.010083649307489395,
0.020225143060088158,
-0.010749172419309616,
0.04452606290578842,
-0.058905210345983505,
-0... |
Tokenization Is More Than Compression | https://aclanthology.org/2024.emnlp-main.40/ | [
"Craig W Schmidt",
"Varshini Reddy",
"Haoran Zhang",
"Alec Alameddine",
"Omri Uzan",
"Yuval Pinter",
"Chris Tanner"
] | Tokenization is a foundational step in natural language processing (NLP) tasks, bridging raw text and language models. Existing tokenization approaches like Byte-Pair Encoding (BPE) originate from the field of data compression, and it has been suggested that the effectiveness of BPE stems from its ability to condense t... | 2024.emnlp-main.40 | 10.18653/v1/2024.emnlp-main.40 | null | 2402.18376 | title_snapshot | [
-0.010740809142589569,
-0.052913472056388855,
-0.02274753898382187,
0.03437834233045578,
0.04688003659248352,
0.033288635313510895,
0.03284038230776787,
0.026143068447709084,
-0.028286907821893692,
-0.03618840500712395,
-0.056440483778715134,
-0.0032787430100142956,
-0.0289242472499609,
0.... |
FLIRT: Feedback Loop In-context Red Teaming | https://aclanthology.org/2024.emnlp-main.41/ | [
"Ninareh Mehrabi",
"Palash Goyal",
"Christophe Dupuy",
"Qian Hu",
"Shalini Ghosh",
"Richard Zemel",
"Kai-Wei Chang",
"Aram Galstyan",
"Rahul Gupta"
] | Warning: this paper contains content that may be inappropriate or offensive.As generative models become available for public use in various applications, testing and analyzing vulnerabilities of these models has become a priority. In this work, we propose an automatic red teaming framework that evaluates a given black-... | 2024.emnlp-main.41 | 10.18653/v1/2024.emnlp-main.41 | null | 2308.04265 | title_snapshot | [
0.014996679499745369,
-0.03532914072275162,
-0.029792537912726402,
0.06385896354913712,
0.006880238186568022,
0.004829966463148594,
0.03800966963171959,
0.0017311765113845468,
-0.017679495736956596,
-0.019544705748558044,
-0.04335777088999748,
0.015437550842761993,
-0.07833012193441391,
-0... |
Successfully Guiding Humans with Imperfect Instructions by Highlighting Potential Errors and Suggesting Corrections | https://aclanthology.org/2024.emnlp-main.42/ | [
"Lingjun Zhao",
"Nguyen X. Khanh",
"Hal Daumé III"
] | Language models will inevitably err in situations with which they are unfamiliar. However, by effectively communicating uncertainties, they can still guide humans toward making sound decisions in those contexts. We demonstrate this idea by developing HEAR, a system that can successfully guide humans in simulated reside... | 2024.emnlp-main.42 | 10.18653/v1/2024.emnlp-main.42 | null | 2402.16973 | title_snapshot | [
-0.0011069687316194177,
0.009676923044025898,
-0.028278548270463943,
0.036222830414772034,
0.05783013999462128,
0.01909877173602581,
0.04026201367378235,
0.04595877230167389,
-0.0339040607213974,
-0.042581092566251755,
-0.07040300220251083,
0.0692048892378807,
-0.04356689378619194,
-0.0198... |
Parameter-Efficient Sparsity Crafting from Dense to Mixture-of-Experts for Instruction Tuning on General Tasks | https://aclanthology.org/2024.emnlp-main.43/ | [
"Haoyuan Wu",
"Haisheng Zheng",
"Zhuolun He",
"Bei Yu"
] | Large language models (LLMs) have demonstrated considerable proficiency in general natural language processing (NLP) tasks. Instruction tuning, a successful paradigm, enhances the ability of LLMs to follow natural language instructions and exhibit robust generalization across general tasks. However, these models often ... | 2024.emnlp-main.43 | 10.18653/v1/2024.emnlp-main.43 | null | 2401.02731 | title_snapshot | [
-0.01939287967979908,
-0.022856729105114937,
-0.02093436010181904,
0.03824422135949135,
0.031101394444704056,
0.05725786089897156,
0.03620944917201996,
0.024874644353985786,
-0.021710002794861794,
-0.020786579698324203,
0.004393396899104118,
0.037156153470277786,
-0.061206940561532974,
0.0... |
GeoGPT4V: Towards Geometric Multi-modal Large Language Models with Geometric Image Generation | https://aclanthology.org/2024.emnlp-main.44/ | [
"Shihao Cai",
"Keqin Bao",
"Hangyu Guo",
"Jizhi Zhang",
"Jun Song",
"Bo Zheng"
] | Large language models have seen widespread adoption in math problem-solving, yet for geometry problems, which often necessitate visual aids even for humans, the most advanced multi-modal models still struggle to effectively utilize image information. High-quality data is crucial for enhancing the geometric capabilities... | 2024.emnlp-main.44 | 10.18653/v1/2024.emnlp-main.44 | null | 2406.11503 | title_snapshot | [
0.0016742453444749117,
-0.021473614498972893,
0.004927510861307383,
0.055312711745500565,
0.008987889625132084,
0.024859949946403503,
0.027895182371139526,
0.05571549013257027,
-0.009857844561338425,
-0.030653279274702072,
-0.022320028394460678,
-0.005158137064427137,
-0.08872400224208832,
... |
DyVo: Dynamic Vocabularies for Learned Sparse Retrieval with Entities | https://aclanthology.org/2024.emnlp-main.45/ | [
"Thong Nguyen",
"Shubham Chatterjee",
"Sean MacAvaney",
"Iain Mackie",
"Jeff Dalton",
"Andrew Yates"
] | Learned Sparse Retrieval (LSR) models use vocabularies from pre-trained transformers, which often split entities into nonsensical fragments. Splitting entities diminishes retrieval accuracy and limits the model’s ability to incorporate up-to-date world knowledge not included in the training data. In this work, we enhan... | 2024.emnlp-main.45 | 10.18653/v1/2024.emnlp-main.45 | null | 2410.07722 | title_snapshot | [
-0.0010444363579154015,
-0.008131103590130806,
-0.008010153658688068,
0.061297085136175156,
0.030806219205260277,
0.014886693097651005,
0.02240517921745777,
0.0034227005671709776,
-0.0301827285438776,
-0.010652323253452778,
-0.005264732521027327,
0.017553770914673805,
-0.07797377556562424,
... |
Let the Expert Stick to His Last: Expert-Specialized Fine-Tuning for Sparse Architectural Large Language Models | https://aclanthology.org/2024.emnlp-main.46/ | [
"Zihan Wang",
"Deli Chen",
"Damai Dai",
"Runxin Xu",
"Zhuoshu Li",
"Yu Wu"
] | Parameter-efficient fine-tuning (PEFT) is crucial for customizing Large Language Models (LLMs) with constrained resource. Although there have been various PEFT methods for dense-architecture LLMs, PEFT for sparse-architecture LLMs is still underexplored. In this work, we study the PEFT method for LLMs with the Mixture-... | 2024.emnlp-main.46 | 10.18653/v1/2024.emnlp-main.46 | null | 2407.01906 | title_snapshot | [
-0.0018906075274571776,
-0.028952855616807938,
0.0029477463103830814,
0.024144111201167107,
0.07133536040782928,
0.03979134559631348,
0.041896753013134,
0.012054077349603176,
-0.018220115453004837,
-0.03742196038365364,
-0.005516084376722574,
0.02450486086308956,
-0.06948629766702652,
-0.0... |
LongEmbed: Extending Embedding Models for Long Context Retrieval | https://aclanthology.org/2024.emnlp-main.47/ | [
"Dawei Zhu",
"Liang Wang",
"Nan Yang",
"Yifan Song",
"Wenhao Wu",
"Furu Wei",
"Sujian Li"
] | Embedding models play a pivotal role in modern NLP applications such as document retrieval. However, existing embedding models are limited to encoding short documents of typically 512 tokens, restrained from application scenarios requiring long inputs. This paper explores context window extension of existing embedding ... | 2024.emnlp-main.47 | 10.18653/v1/2024.emnlp-main.47 | null | 2404.12096 | title_snapshot | [
-0.02445387654006481,
-0.05877236649394035,
-0.02645840123295784,
0.025169968605041504,
0.033382073044776917,
0.005021058022975922,
0.012524359859526157,
0.002232685685157776,
-0.027299879118800163,
-0.03037928231060505,
-0.057897571474313736,
0.015010254457592964,
-0.040580954402685165,
-... |
Making Large Language Models Better Reasoners with Orchestrated Streaming Experiences | https://aclanthology.org/2024.emnlp-main.48/ | [
"Xiangyang Liu",
"Junliang He",
"Xipeng Qiu"
] | Large language models (LLMs) can perform complex reasoning by generating intermediate reasoning steps using chain-of-thought prompting under zero-shot or few-shot settings. However, zero-shot prompting always encounters low performance, and the superior performance of few-shot prompting hinges on the manual-crafting of... | 2024.emnlp-main.48 | 10.18653/v1/2024.emnlp-main.48 | null | 2504.00473 | title_snapshot | [
-0.018354305997490883,
-0.028055157512426376,
0.008151175454258919,
0.05295603349804878,
0.04297855496406555,
0.01235928200185299,
0.012504106387495995,
0.005636671092361212,
-0.02265181578695774,
0.010097863152623177,
-0.01673695631325245,
0.03428327292203903,
-0.05734121426939964,
-0.006... |
Overcome Noise and Bias: Segmentation-Aided Multi-Granularity Denoising and Debiasing for Enhanced Quarduples Extraction in Dialogue | https://aclanthology.org/2024.emnlp-main.49/ | [
"Xianlong Luo",
"Meng Yang",
"Yihao Wang"
] | Dialogue Aspect-based Sentiment Quadruple analysis (DiaASQ) extends ABSA to more complex real-world scenarios (i.e., dialogues), which makes existing generation methods encounter heightened noise and order bias challenges, leading to decreased robustness and accuracy.To address these, we propose the Segmentation-Aided ... | 2024.emnlp-main.49 | 10.18653/v1/2024.emnlp-main.49 | null | null | null | [
-0.0259652528911829,
-0.008744128979742527,
-0.005423806142061949,
0.05425102263689041,
-0.007193929050117731,
0.06842442601919174,
0.02854004316031933,
0.00679340586066246,
-0.025613699108362198,
-0.04465740919113159,
-0.03092915005981922,
0.04109041392803192,
-0.08258818835020065,
0.0078... |
Integrating Plutchik’s Theory with Mixture of Experts for Enhancing Emotion Classification | https://aclanthology.org/2024.emnlp-main.50/ | [
"Dongjun Lim",
"Yun-Gyung Cheong"
] | Emotion significantly influences human behavior and decision-making processes. We propose a labeling methodology grounded in Plutchik’s Wheel of Emotions theory for emotion classification. Furthermore, we employ a Mixture of Experts (MoE) architecture to evaluate the efficacy of this labeling approach, by identifying t... | 2024.emnlp-main.50 | 10.18653/v1/2024.emnlp-main.50 | null | null | null | [
-0.033253345638513565,
-0.02533092349767685,
-0.022661857306957245,
0.03755342960357666,
-0.011127663776278496,
0.018105704337358475,
0.02623838372528553,
0.011274473741650581,
-0.030284782871603966,
-0.029299700632691383,
-0.02703288197517395,
0.04355887696146965,
-0.06509454548358917,
-0... |
In-context Contrastive Learning for Event Causality Identification | https://aclanthology.org/2024.emnlp-main.51/ | [
"Liang Chao",
"Wei Xiang",
"Bang Wang"
] | Event Causality Identification (ECI) aims at determining the existence of a causal relation between two events. Although recent prompt learning-based approaches have shown promising improvements on the ECI task, their performance are often subject to the delicate design of multiple prompts and the positive correlations... | 2024.emnlp-main.51 | 10.18653/v1/2024.emnlp-main.51 | null | 2405.10512 | title_snapshot | [
-0.011000133119523525,
-0.020753294229507446,
-0.02006242424249649,
0.039772167801856995,
0.005653051659464836,
0.006219572387635708,
0.023803183808922768,
0.028227297589182854,
-0.032693468034267426,
0.011804214678704739,
-0.006824617739766836,
0.04563011974096298,
-0.04136338084936142,
0... |
What’s Mine becomes Yours: Defining, Annotating and Detecting Context-Dependent Paraphrases in News Interview Dialogs | https://aclanthology.org/2024.emnlp-main.52/ | [
"Anna Wegmann",
"Tijs A. Van Den Broek",
"Dong Nguyen"
] | Best practices for high conflict conversations like counseling or customer support almost always include recommendations to paraphrase the previous speaker. Although paraphrase classification has received widespread attention in NLP, paraphrases are usually considered independent from context, and common models and dat... | 2024.emnlp-main.52 | 10.18653/v1/2024.emnlp-main.52 | null | 2404.06670 | title_snapshot | [
-0.027563568204641342,
-0.00008564261952415109,
-0.015659326687455177,
0.02497629076242447,
0.06207139417529106,
0.031019140034914017,
0.012982108630239964,
0.016379570588469505,
-0.0011143481824547052,
-0.02192620001733303,
-0.07180846482515335,
0.05701925978064537,
-0.03599818795919418,
... |
Language Models Learn Rare Phenomena from Less Rare Phenomena: The Case of the Missing AANNs | https://aclanthology.org/2024.emnlp-main.53/ | [
"Kanishka Misra",
"Kyle Mahowald"
] | Language models learn rare syntactic phenomena, but the extent to which this is attributable to generalization vs. memorization is a major open question. To that end, we iteratively trained transformer language models on systematically manipulated corpora which were human-scale in size, and then evaluated their learnin... | 2024.emnlp-main.53 | 10.18653/v1/2024.emnlp-main.53 | Outstanding Paper Award | 2403.19827 | title_snapshot | [
-0.01114457007497549,
-0.01243832428008318,
-0.011111143045127392,
0.019291182979941368,
0.0318758487701416,
0.0006636488251388073,
0.05088655278086662,
0.03127897158265114,
-0.04489169269800186,
0.0086765605956316,
-0.02146291919052601,
0.04864869266748428,
-0.05154789611697197,
0.0091821... |
Large Language Models for Data Annotation and Synthesis: A Survey | https://aclanthology.org/2024.emnlp-main.54/ | [
"Zhen Tan",
"Dawei Li",
"Song Wang",
"Alimohammad Beigi",
"Bohan Jiang",
"Amrita Bhattacharjee",
"Mansooreh Karami",
"Jundong Li",
"Lu Cheng",
"Huan Liu"
] | Data annotation and synthesis generally refers to the labeling or generating of raw data with relevant information, which could be used for improving the efficacy of machine learning models. The process, however, is labor-intensive and costly. The emergence of advanced Large Language Models (LLMs), exemplified by GPT-4... | 2024.emnlp-main.54 | 10.18653/v1/2024.emnlp-main.54 | null | 2402.13446 | title_snapshot | [
-0.025414882227778435,
-0.04511851444840431,
-0.014885942451655865,
0.02246207743883133,
0.05071946606040001,
0.01808144338428974,
0.015078060328960419,
0.016130076721310616,
-0.023175349459052086,
0.008185168728232384,
-0.03824654221534729,
0.03472711890935898,
-0.062171660363674164,
0.01... |
Chain-of-Dictionary Prompting Elicits Translation in Large Language Models | https://aclanthology.org/2024.emnlp-main.55/ | [
"Hongyuan Lu",
"Haoran Yang",
"Haoyang Huang",
"Dongdong Zhang",
"Wai Lam",
"Furu Wei"
] | Large language models (LLMs) have shown surprisingly good performance in multilingual neural machine translation (MNMT) even if not being trained explicitly for translation. Yet, they still struggle with translating low-resource languages. As supported by our experiments, a bilingual dictionary between the source and t... | 2024.emnlp-main.55 | 10.18653/v1/2024.emnlp-main.55 | null | 2305.06575 | title_snapshot | [
-0.0164941418915987,
-0.05014133080840111,
-0.008645212277770042,
0.04384085163474083,
0.01611418090760708,
0.027862219139933586,
0.018642669543623924,
0.02145528234541416,
-0.014526418410241604,
0.004787682089954615,
-0.03913295269012451,
0.04789989814162254,
-0.06311944872140884,
-0.0169... |
AdaZeta: Adaptive Zeroth-Order Tensor-Train Adaption for Memory-Efficient Large Language Models Fine-Tuning | https://aclanthology.org/2024.emnlp-main.56/ | [
"Yifan Yang",
"Kai Zhen",
"Ershad Banijamali",
"Athanasios Mouchtaris",
"Zheng Zhang"
] | Fine-tuning large language models (LLMs) has achieved remarkable performance across various natural language processing tasks, yet it demands more and more memory as model sizes keep growing. To address this issue, the recently proposed Memory-efficient Zeroth-order (MeZO) methods attempt to fine-tune LLMs using only f... | 2024.emnlp-main.56 | 10.18653/v1/2024.emnlp-main.56 | null | 2406.18060 | title_snapshot | [
-0.04938464239239693,
-0.0349399596452713,
-0.01210120227187872,
0.041726112365722656,
0.033215176314115524,
0.02984793484210968,
0.04107225313782692,
0.03267134726047516,
-0.028868475928902626,
0.006615354213863611,
-0.0021625636145472527,
0.046881455928087234,
-0.05490301176905632,
0.017... |
RoseLoRA: Row and Column-wise Sparse Low-rank Adaptation of Pre-trained Language Model for Knowledge Editing and Fine-tuning | https://aclanthology.org/2024.emnlp-main.57/ | [
"Haoyu Wang",
"Tianci Liu",
"Ruirui Li",
"Monica Xiao Cheng",
"Tuo Zhao",
"Jing Gao"
] | Pre-trained language models, trained on large-scale corpora, demonstrate strong generalizability across various NLP tasks. Fine-tuning these models for specific tasks typically involves updating all parameters, which is resource-intensive. Parameter-efficient fine-tuning (PEFT) methods, such as the popular LoRA family,... | 2024.emnlp-main.57 | 10.18653/v1/2024.emnlp-main.57 | null | 2406.10777 | title_snapshot | [
-0.021350963041186333,
-0.03499848023056984,
0.0028824545443058014,
0.04821763560175896,
0.0371699258685112,
0.02828553318977356,
0.03504415974020958,
0.005539924371987581,
-0.01694624312222004,
0.009209468960762024,
-0.006067855749279261,
0.023163622245192528,
-0.08008144795894623,
-0.020... |
BlendFilter: Advancing Retrieval-Augmented Large Language Models via Query Generation Blending and Knowledge Filtering | https://aclanthology.org/2024.emnlp-main.58/ | [
"Haoyu Wang",
"Ruirui Li",
"Haoming Jiang",
"Jinjin Tian",
"Zhengyang Wang",
"Chen Luo",
"Xianfeng Tang",
"Monica Xiao Cheng",
"Tuo Zhao",
"Jing Gao"
] | Retrieval-augmented Large Language Models (LLMs) offer substantial benefits in enhancing performance across knowledge-intensive scenarios. However, these methods often struggle with complex inputs and encounter difficulties due to noisy knowledge retrieval, notably hindering model effectiveness. To address this issue, ... | 2024.emnlp-main.58 | 10.18653/v1/2024.emnlp-main.58 | null | 2402.11129 | title_snapshot | [
0.0037204851396381855,
-0.02754085138440132,
0.003939129877835512,
0.027691414579749107,
0.049848251044750214,
-0.017881792038679123,
0.030686665326356888,
-0.0053927479311823845,
-0.03973771631717682,
0.008507480844855309,
-0.052392084151506424,
0.0308313500136137,
-0.058460719883441925,
... |
HEART-felt Narratives: Tracing Empathy and Narrative Style in Personal Stories with LLMs | https://aclanthology.org/2024.emnlp-main.59/ | [
"Jocelyn Shen",
"Joel Mire",
"Hae Won Park",
"Cynthia Breazeal",
"Maarten Sap"
] | Empathy serves as a cornerstone in enabling prosocial behaviors, and can be evoked through sharing of personal experiences in stories. While empathy is influenced by narrative content, intuitively, people respond to the way a story is told as well, through narrative style. Yet the relationship between empathy and narra... | 2024.emnlp-main.59 | 10.18653/v1/2024.emnlp-main.59 | null | 2405.17633 | title_snapshot | [
-0.0033919084817171097,
-0.03545431047677994,
-0.009836075827479362,
-0.007907639257609844,
0.04598184674978256,
0.012572425417602062,
0.014112592674791813,
0.011639002710580826,
-0.0009577999589964747,
-0.02213580347597599,
-0.027877219021320343,
0.000725218968000263,
-0.036632005125284195,... |
Eliminating Biased Length Reliance of Direct Preference Optimization via Down-Sampled KL Divergence | https://aclanthology.org/2024.emnlp-main.60/ | [
"Junru Lu",
"Jiazheng Li",
"Siyu An",
"Meng Zhao",
"Yulan He",
"Di Yin",
"Xing Sun"
] | Direct Preference Optimization (DPO) has emerged as a prominent algorithm for the direct and robust alignment of Large Language Models (LLMs) with human preferences, offering a more straightforward alternative to the complex Reinforcement Learning from Human Feedback (RLHF). Despite its promising efficacy, DPO faces a ... | 2024.emnlp-main.60 | 10.18653/v1/2024.emnlp-main.60 | null | 2406.10957 | title_snapshot | [
-0.03388923779129982,
-0.013543075881898403,
0.004860404878854752,
0.055790551006793976,
0.028175899758934975,
0.02572888880968094,
0.024580422788858414,
0.03107578307390213,
-0.02396254800260067,
-0.01775052770972252,
-0.01729479245841503,
0.038857102394104004,
-0.08906615525484085,
-0.00... |
Bridging Cultures in the Kitchen: A Framework and Benchmark for Cross-Cultural Recipe Retrieval | https://aclanthology.org/2024.emnlp-main.61/ | [
"Tianyi Hu",
"Maria Maistro",
"Daniel Hershcovich"
] | The cross-cultural adaptation of recipes is an important application of identifying and bridging cultural differences in language. The challenge lies in retaining the essence of the original recipe while also aligning with the writing and dietary habits of the target culture. Information Retrieval (IR) offers a way to ... | 2024.emnlp-main.61 | 10.18653/v1/2024.emnlp-main.61 | null | null | null | [
-0.018302734941244125,
0.017430823296308517,
-0.04347338154911995,
0.019740911200642586,
0.044204019010066986,
-0.012689832597970963,
0.010888720862567425,
0.04210186377167702,
-0.03308169171214104,
-0.020430896431207657,
-0.025786621496081352,
0.02983739785850048,
-0.021219352260231972,
-... |
RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models | https://aclanthology.org/2024.emnlp-main.62/ | [
"Peng Xia",
"Kangyu Zhu",
"Haoran Li",
"Hongtu Zhu",
"Yun Li",
"Gang Li",
"Linjun Zhang",
"Huaxiu Yao"
] | The recent emergence of Medical Large Vision Language Models (Med-LVLMs) has enhanced medical diagnosis. However, current Med-LVLMs frequently encounter factual issues, often generating responses that do not align with established medical facts. Retrieval-Augmented Generation (RAG), which utilizes external knowledge, c... | 2024.emnlp-main.62 | 10.18653/v1/2024.emnlp-main.62 | null | 2407.05131 | title_snapshot | [
0.010871262289583683,
-0.011420338414609432,
0.02018192782998085,
0.03561544045805931,
0.02135643921792507,
-0.03197397664189339,
0.03623013570904732,
-0.0021761436946690083,
-0.03745139017701149,
-0.01685899682343006,
-0.05738421529531479,
0.06330091506242752,
-0.08004476875066757,
0.0094... |
CryptoTrade: A Reflective LLM-based Agent to Guide Zero-shot Cryptocurrency Trading | https://aclanthology.org/2024.emnlp-main.63/ | [
"Yuan Li",
"Bingqiao Luo",
"Qian Wang",
"Nuo Chen",
"Xu Liu",
"Bingsheng He"
] | The utilization of Large Language Models (LLMs) in financial trading has primarily been concentrated within the stock market, aiding in economic and financial decisions. Yet, the unique opportunities presented by the cryptocurrency market, noted for its on-chain data’s transparency and the critical influence of off-cha... | 2024.emnlp-main.63 | 10.18653/v1/2024.emnlp-main.63 | null | 2407.09546 | title_judge | [
0.01394446101039648,
-0.05842737480998039,
-0.014944924972951412,
0.031102711334824562,
0.055240094661712646,
0.008727803826332092,
0.0054837786592543125,
0.04700177162885666,
-0.019475724548101425,
-0.013332782313227654,
-0.021362414583563805,
0.014055266976356506,
-0.04969537630677223,
0... |
A Survey on In-context Learning | https://aclanthology.org/2024.emnlp-main.64/ | [
"Qingxiu Dong",
"Lei Li",
"Damai Dai",
"Ce Zheng",
"Jingyuan Ma",
"Rui Li",
"Heming Xia",
"Jingjing Xu",
"Zhiyong Wu",
"Baobao Chang",
"Xu Sun",
"Lei Li",
"Zhifang Sui"
] | With the increasing capabilities of large language models (LLMs), in-context learning (ICL) has emerged as a new paradigm for natural language processing (NLP), where LLMs make predictions based on contexts augmented with a few examples. It has been a significant trend to explore ICL to evaluate and extrapolate the abi... | 2024.emnlp-main.64 | 10.18653/v1/2024.emnlp-main.64 | null | 2301.00234 | title_snapshot | [
-0.015460819005966187,
0.0032768910750746727,
-0.020802680402994156,
0.017237424850463867,
0.030400602146983147,
-0.002612857846543193,
0.022319775074720383,
0.0424339734017849,
-0.02710159681737423,
0.020968209952116013,
-0.047965262085199356,
0.04781286418437958,
-0.03841806575655937,
-0... |
DocHieNet: A Large and Diverse Dataset for Document Hierarchy Parsing | https://aclanthology.org/2024.emnlp-main.65/ | [
"Hangdi Xing",
"Changxu Cheng",
"Feiyu Gao",
"Zirui Shao",
"Zhi Yu",
"Jiajun Bu",
"Qi Zheng",
"Cong Yao"
] | Parsing documents from pixels, such as pictures and scanned PDFs, into hierarchical structures is extensively demanded in the daily routines of data storage, retrieval and understanding. However, previously the research on this topic has been largely hindered since most existing datasets are small-scale, or contain doc... | 2024.emnlp-main.65 | 10.18653/v1/2024.emnlp-main.65 | null | null | null | [
-0.03574735298752785,
-0.016361979767680168,
-0.02183869667351246,
0.054150328040122986,
0.016567613929510117,
-0.006465605925768614,
0.012294718995690346,
-0.010108300484716892,
-0.020327571779489517,
-0.05804013833403587,
-0.021199936047196388,
0.004420462064445019,
-0.056303828954696655,
... |
AMR-Evol: Adaptive Modular Response Evolution Elicits Better Knowledge Distillation for Large Language Models in Code Generation | https://aclanthology.org/2024.emnlp-main.66/ | [
"Ziyang Luo",
"Xin Li",
"Hongzhan Lin",
"Jing Ma",
"Lidong Bing"
] | The impressive performance of proprietary LLMs like GPT4 in code generation has led to a trend to replicate these capabilities in open-source models through knowledge distillation (e.g. Code Evol-Instruct). However, these efforts often neglect the crucial aspect of response quality, relying heavily on teacher models fo... | 2024.emnlp-main.66 | 10.18653/v1/2024.emnlp-main.66 | null | 2410.00558 | title_snapshot | [
0.010599937289953232,
-0.03934653103351593,
-0.01121172122657299,
0.03015240468084812,
0.04817941412329674,
0.020470816642045975,
0.054702334105968475,
-0.022931218147277832,
-0.025409644469618797,
-0.035816971212625504,
-0.027383718639612198,
0.014071465469896793,
-0.08420700579881668,
-0... |
EFUF: Efficient Fine-Grained Unlearning Framework for Mitigating Hallucinations in Multimodal Large Language Models | https://aclanthology.org/2024.emnlp-main.67/ | [
"Shangyu Xing",
"Fei Zhao",
"Zhen Wu",
"Tuo An",
"Weihao Chen",
"Chunhui Li",
"Jianbing Zhang",
"Xinyu Dai"
] | Multimodal large language models (MLLMs) have attracted increasing attention in the past few years, but they may still generate descriptions that include objects not present in the corresponding images, a phenomenon known as object hallucination. To eliminate hallucinations, existing methods manually annotate paired re... | 2024.emnlp-main.67 | 10.18653/v1/2024.emnlp-main.67 | null | 2402.09801 | title_snapshot | [
0.04290812462568283,
-0.006371357478201389,
0.029969705268740654,
0.015747247263789177,
0.023898961022496223,
-0.005830997135490179,
0.05020559951663017,
0.034522317349910736,
-0.025693664327263832,
-0.036538492888212204,
-0.01759512722492218,
0.02502080239355564,
-0.0899992436170578,
-0.0... |
Rethinking Pruning Large Language Models: Benefits and Pitfalls of Reconstruction Error Minimization | https://aclanthology.org/2024.emnlp-main.68/ | [
"Sungbin Shin",
"Wonpyo Park",
"Jaeho Lee",
"Namhoon Lee"
] | This work suggests fundamentally rethinking the current practice of pruning large language models (LLMs). The way it is done is by divide and conquer: split the model into submodels, sequentially prune them, and reconstruct predictions of the dense counterparts on small calibration data one at a time; the final model i... | 2024.emnlp-main.68 | 10.18653/v1/2024.emnlp-main.68 | null | 2406.15524 | title_snapshot | [
-0.016780100762844086,
-0.011360817588865757,
-0.035141851752996445,
0.03211355209350586,
0.051380280405282974,
0.03965804725885391,
0.02228350006043911,
0.012488359585404396,
-0.03206977993249893,
-0.019750049337744713,
-0.011735864914953709,
0.04438425227999687,
-0.047374945133924484,
0.... |
LLMs Are Zero-Shot Context-Aware Simultaneous Translators | https://aclanthology.org/2024.emnlp-main.69/ | [
"Roman Koshkin",
"Katsuhito Sudoh",
"Satoshi Nakamura"
] | The advent of transformers has fueled progress in machine translation. More recently large language models (LLMs) have come to the spotlight thanks to their generality and strong performance in a wide range of language tasks, including translation. Here we show that open-source LLMs perform on par with or better than s... | 2024.emnlp-main.69 | 10.18653/v1/2024.emnlp-main.69 | null | 2406.13476 | title_snapshot | [
-0.008797794580459595,
-0.022653469815850258,
-0.0006019118009135127,
0.015787165611982346,
0.04870716109871864,
0.037152018398046494,
0.03153418004512787,
0.05463460087776184,
-0.01509818620979786,
0.015097050927579403,
-0.03818352147936821,
0.05950642749667168,
-0.05377965047955513,
-0.0... |
AgentReview: Exploring Peer Review Dynamics with LLM Agents | https://aclanthology.org/2024.emnlp-main.70/ | [
"Yiqiao Jin",
"Qinlin Zhao",
"Yiyang Wang",
"Hao Chen",
"Kaijie Zhu",
"Yijia Xiao",
"Jindong Wang"
] | Peer review is fundamental to the integrity and advancement of scientific publication. Traditional methods of peer review analyses often rely on exploration and statistics of existing peer review data, which do not adequately address the multivariate nature of the process, account for the latent variables, and are furt... | 2024.emnlp-main.70 | 10.18653/v1/2024.emnlp-main.70 | null | 2406.12708 | title_snapshot | [
-0.005679877009242773,
-0.022189194336533546,
-0.017800096422433853,
0.02244725078344345,
0.03847496584057808,
-0.038931239396333694,
0.03481029346585274,
-0.0000425013349740766,
-0.011600418947637081,
-0.029949426651000977,
-0.0101394047960639,
0.03361042961478233,
-0.04154110699892044,
-... |
ChatRetriever: Adapting Large Language Models for Generalized and Robust Conversational Dense Retrieval | https://aclanthology.org/2024.emnlp-main.71/ | [
"Kelong Mao",
"Chenlong Deng",
"Haonan Chen",
"Fengran Mo",
"Zheng Liu",
"Tetsuya Sakai",
"Zhicheng Dou"
] | Conversational search requires accurate interpretation of user intent from complex multi-turn contexts. This paper presents ChatRetriever, which inherits the strong generalization capability of large language models to robustly represent complex conversational sessions for dense retrieval. To achieve this, we propose a... | 2024.emnlp-main.71 | 10.18653/v1/2024.emnlp-main.71 | null | 2404.13556 | title_snapshot | [
-0.026864830404520035,
-0.055844660848379135,
-0.013906905427575111,
0.043139632791280746,
0.022917388007044792,
-0.007380319759249687,
0.006426872685551643,
0.031241407617926598,
-0.012282591313123703,
-0.0002698110183700919,
-0.0399949885904789,
0.04761843383312225,
-0.03767155483365059,
... |
Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgments | https://aclanthology.org/2024.emnlp-main.72/ | [
"Han Zhou",
"Xingchen Wan",
"Yinhong Liu",
"Nigel Collier",
"Ivan Vulić",
"Anna Korhonen"
] | Large language models (LLMs) have shown promising abilities as cost-effective and reference-free evaluators for assessing language generation quality. In particular, pairwise LLM evaluators, which compare two generated texts and determine the preferred one, have been employed in a wide range of applications. However, L... | 2024.emnlp-main.72 | 10.18653/v1/2024.emnlp-main.72 | null | 2406.11370 | title_snapshot | [
-0.0037742580752819777,
0.002679358934983611,
-0.010282271541655064,
0.035792190581560135,
0.02603021264076233,
0.028339775279164314,
0.002317957114428282,
0.049657054245471954,
-0.021842999383807182,
-0.010547329671680927,
-0.016443578526377678,
0.04326551780104637,
-0.06891103833913803,
... |
Learning Interpretable Legal Case Retrieval via Knowledge-Guided Case Reformulation | https://aclanthology.org/2024.emnlp-main.73/ | [
"Chenlong Deng",
"Kelong Mao",
"Zhicheng Dou"
] | Legal case retrieval for sourcing similar cases is critical in upholding judicial fairness. Different from general web search, legal case retrieval involves processing lengthy, complex, and highly specialized legal documents. Existing methods in this domain often overlook the incorporation of legal expert knowledge, wh... | 2024.emnlp-main.73 | 10.18653/v1/2024.emnlp-main.73 | null | 2406.19760 | title_snapshot | [
-0.034302763640880585,
-0.0451008565723896,
-0.011659367941319942,
0.023904120549559593,
0.047164250165224075,
-0.022502949461340904,
0.03378688171505928,
-0.01870940998196602,
-0.019912075251340866,
-0.014861579984426498,
-0.022194741293787956,
0.04005441442131996,
-0.04571346938610077,
0... |
Effective Demonstration Annotation for In-Context Learning via Language Model-Based Determinantal Point Process | https://aclanthology.org/2024.emnlp-main.74/ | [
"Peng Wang",
"Xiaobin Wang",
"Chao Lou",
"Shengyu Mao",
"Pengjun Xie",
"Yong Jiang"
] | In-context learning (ICL) is a few-shot learning paradigm that involves learning mappings through input-output pairs and appropriately applying them to new instances. Despite the remarkable ICL capabilities demonstrated by Large Language Models (LLMs), existing works are highly dependent on large-scale labeled support ... | 2024.emnlp-main.74 | 10.18653/v1/2024.emnlp-main.74 | null | 2408.02103 | title_snapshot | [
-0.017630835995078087,
0.0045556421391665936,
-0.021032152697443962,
0.026173489168286324,
0.017862919718027115,
0.024912318214774132,
0.03531777486205101,
0.029629886150360107,
-0.028104916214942932,
0.01801690272986889,
-0.03556700423359871,
0.03188302367925644,
-0.05052653327584267,
-0.... |
Pre-trained Language Models Do Not Help Auto-regressive Text-to-Image Generation | https://aclanthology.org/2024.emnlp-main.75/ | [
"Yuhui Zhang",
"Brandon McKinzie",
"Zhe Gan",
"Vaishaal Shankar",
"Alexander T Toshev"
] | Recent advances in image tokenizers, such as VQ-VAE, have enabled text-to-image generation using auto-regressive methods, similar to language modeling. However, these methods have yet to leverage pre-trained language models, despite their adaptability to various downstream tasks. In this work, we explore this gap by ad... | 2024.emnlp-main.75 | 10.18653/v1/2024.emnlp-main.75 | null | 2311.16201 | title_snapshot | [
0.007180043496191502,
-0.030106792226433754,
-0.011032951064407825,
0.05121660232543945,
0.031302470713853836,
0.04502139985561371,
0.046226173639297485,
0.02059289813041687,
-0.019714966416358948,
-0.011879989877343178,
-0.05280671268701553,
0.009142235852777958,
-0.042913246899843216,
0.... |
QUDSELECT: Selective Decoding for Questions Under Discussion Parsing | https://aclanthology.org/2024.emnlp-main.76/ | [
"Ashima Suvarna",
"Xiao Liu",
"Tanmay Parekh",
"Kai-Wei Chang",
"Nanyun Peng"
] | Question Under Discussion (QUD) is a discourse framework that uses implicit questions to reveal discourse relationships between sentences. In QUD parsing, each sentence is viewed as an answer to a question triggered by an anchor sentence in prior context. The resulting QUD structure is required to conform to several th... | 2024.emnlp-main.76 | 10.18653/v1/2024.emnlp-main.76 | null | 2408.01046 | title_snapshot | [
-0.019399486482143402,
-0.014627459459006786,
-0.03538324683904648,
0.05198805406689644,
0.031901825219392776,
0.04436906427145004,
0.01301935687661171,
0.02150055766105652,
-0.010698061436414719,
-0.005078430753201246,
-0.03577379882335663,
0.04589930176734924,
-0.07036162912845612,
-0.00... |
Mitigating Language Bias of LMMs in Social Intelligence Understanding with Virtual Counterfactual Calibration | https://aclanthology.org/2024.emnlp-main.77/ | [
"Peng Chen",
"Xiao-Yu Guo",
"Yuan-Fang Li",
"Xiaowang Zhang",
"Zhiyong Feng"
] | null | 2024.emnlp-main.77 | 10.18653/v1/2024.emnlp-main.77 | null | null | null | [
0.03727981075644493,
-0.00928017683327198,
-0.00822213664650917,
0.02351911924779415,
0.016701772809028625,
0.03465837240219116,
0.07918785512447357,
0.02553977444767952,
-0.02950289659202099,
-0.0036478627007454634,
0.0006956046563573182,
0.07077974081039429,
-0.06044696643948555,
-0.0240... |
Model Balancing Helps Low-data Training and Fine-tuning | https://aclanthology.org/2024.emnlp-main.78/ | [
"Zihang Liu",
"Yuanzhe Hu",
"Tianyu Pang",
"Yefan Zhou",
"Pu Ren",
"Yaoqing Yang"
] | Recent advances in foundation models have emphasized the need to align pre-trained models with specialized domains using small, curated datasets. Studies on these foundation models underscore the importance of low-data training and fine-tuning. This topic, well-known in natural language processing (NLP), has also gaine... | 2024.emnlp-main.78 | 10.18653/v1/2024.emnlp-main.78 | null | 2410.12178 | title_snapshot | [
0.003166796639561653,
-0.04956890270113945,
-0.004376026801764965,
0.026822861284017563,
0.057465799152851105,
-0.001406516064889729,
0.025829486548900604,
-0.005612816661596298,
-0.02002301625907421,
-0.051925040781497955,
-0.017855048179626465,
0.02093246392905712,
-0.06976045668125153,
... |
Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment | https://aclanthology.org/2024.emnlp-main.79/ | [
"Zhaofeng Wu",
"Ananth Balashankar",
"Yoon Kim",
"Jacob Eisenstein",
"Ahmad Beirami"
] | Aligning language models (LMs) based on human-annotated preference data is a crucial step in obtaining practical and performant LM-based systems. However, multilingual human preference data are difficult to obtain at scale, making it challenging to extend this framework to diverse languages. In this work, we evaluate a... | 2024.emnlp-main.79 | 10.18653/v1/2024.emnlp-main.79 | null | 2404.12318 | title_snapshot | [
0.005315827205777168,
0.0031630874145776033,
-0.011233721859753132,
0.031906139105558395,
0.04058333858847618,
0.029260575771331787,
0.013148167170584202,
0.04649621248245239,
0.005834274459630251,
-0.03060321882367134,
-0.04273366183042526,
0.05407819524407387,
-0.07588409632444382,
-0.03... |
Large Language Models as Foundations for Next-Gen Dense Retrieval: A Comprehensive Empirical Assessment | https://aclanthology.org/2024.emnlp-main.80/ | [
"Kun Luo",
"Minghao Qin",
"Zheng Liu",
"Shitao Xiao",
"Jun Zhao",
"Kang Liu"
] | Pre-trained language models like BERT and T5 serve as crucial backbone encoders for dense retrieval. However, these models often exhibit limited generalization capabilities and face challenges in improving in-domain accuracy. Recent research has explored using large language models (LLMs) as retrievers, achieving state... | 2024.emnlp-main.80 | 10.18653/v1/2024.emnlp-main.80 | null | 2408.12194 | title_snapshot | [
-0.028233619406819344,
-0.03738964721560478,
-0.002927616238594055,
0.03910679742693901,
0.028454700484871864,
-0.01162588968873024,
0.01533753052353859,
0.031649746000766754,
-0.00409072358161211,
0.00860923808068037,
-0.029957247897982597,
0.04593314230442047,
-0.05486077070236206,
0.023... |
A New Pipeline for Knowledge Graph Reasoning Enhanced by Large Language Models Without Fine-Tuning | https://aclanthology.org/2024.emnlp-main.81/ | [
"Zhongwu Chen",
"Long Bai",
"Zixuan Li",
"Zhen Huang",
"Xiaolong Jin",
"Yong Dou"
] | Conventional Knowledge Graph Reasoning (KGR) models learn the embeddings of KG components over the structure of KGs, but their performances are limited when the KGs are severely incomplete. Recent LLM-enhanced KGR models input KG structural information into LLMs. However, they require fine-tuning on open-source LLMs an... | 2024.emnlp-main.81 | 10.18653/v1/2024.emnlp-main.81 | null | null | null | [
0.0008634152472950518,
-0.018903587013483047,
0.025266597047448158,
0.04407186433672905,
0.05112204700708389,
0.011413095518946648,
0.03471805527806282,
-0.014182733371853828,
0.006062732543796301,
0.011636915616691113,
-0.007504028268158436,
0.038933850824832916,
-0.050221990793943405,
0.... |
Towards Tool Use Alignment of Large Language Models | https://aclanthology.org/2024.emnlp-main.82/ | [
"Zhi-Yuan Chen",
"Shiqi Shen",
"Guangyao Shen",
"Gong Zhi",
"Xu Chen",
"Yankai Lin"
] | Recently, tool use with LLMs has become one of the primary research topics as it can help LLM generate truthful and helpful responses. Existing studies on tool use with LLMs primarily focus on enhancing the tool-calling ability of LLMs. In practice, like chat assistants, LLMs are also required to align with human value... | 2024.emnlp-main.82 | 10.18653/v1/2024.emnlp-main.82 | null | null | null | [
0.008236121386289597,
0.006539495661854744,
-0.041475433856248856,
0.009450222365558147,
0.03945508226752281,
0.021145375445485115,
0.03338241949677467,
0.012430716305971146,
-0.009962677955627441,
-0.023782160133123398,
-0.03600174933671951,
0.04067583382129669,
-0.08258553594350815,
-0.0... |
DecorateLM: Data Engineering through Corpus Rating, Tagging, and Editing with Language Models | https://aclanthology.org/2024.emnlp-main.83/ | [
"Ranchi Zhao",
"Zhen Leng Thai",
"Yifan Zhang",
"Shengding Hu",
"Jie Zhou",
"Yunqi Ba",
"Jie Cai",
"Zhiyuan Liu",
"Maosong Sun"
] | The performance of Large Language Models (LLMs) is substantially influenced by the pretraining corpus, which consists of vast quantities of unsupervised data processed by the models. Despite its critical role in model performance, ensuring the quality of this data is challenging due to its sheer volume and the absence ... | 2024.emnlp-main.83 | 10.18653/v1/2024.emnlp-main.83 | null | 2410.05639 | title_snapshot | [
0.019382262602448463,
-0.021574176847934723,
0.008072610944509506,
0.02582443691790104,
0.05067305266857147,
0.004036750644445419,
0.023235315456986427,
-0.013945754617452621,
-0.034273263067007065,
0.0033928423654288054,
-0.05402692034840584,
0.03288285806775093,
-0.0646573007106781,
-0.0... |
Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps | https://aclanthology.org/2024.emnlp-main.84/ | [
"Yung-Sung Chuang",
"Linlu Qiu",
"Cheng-Yu Hsieh",
"Ranjay Krishna",
"Yoon Kim",
"James R. Glass"
] | When asked to summarize articles or answer questions given a passage, large language models (LLMs) can hallucinate details and respond with unsubstantiated answers that are inaccurate with respect to the input context. This paper describes a simple approach for detecting such **contextual hallucinations**. We hypothesi... | 2024.emnlp-main.84 | 10.18653/v1/2024.emnlp-main.84 | null | 2407.07071 | title_snapshot | [
-0.0018023357260972261,
0.024906296283006668,
0.020080246031284332,
0.012003547511994839,
0.019581209868192673,
0.02034868858754635,
0.04935111477971077,
0.0406915582716465,
-0.03256601467728615,
-0.03497776389122009,
-0.057419553399086,
0.04449726641178131,
-0.049997296184301376,
-0.00967... |
Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment | https://aclanthology.org/2024.emnlp-main.85/ | [
"Yiju Guo",
"Ganqu Cui",
"Lifan Yuan",
"Ning Ding",
"Zexu Sun",
"Bowen Sun",
"Huimin Chen",
"Ruobing Xie",
"Jie Zhou",
"Yankai Lin",
"Zhiyuan Liu",
"Maosong Sun"
] | Alignment in artificial intelligence pursues the consistency between model responses and human preferences as well as values. In practice, the multifaceted nature of human preferences inadvertently introduces what is known as the ”alignment tax”–a compromise where enhancements in alignment within one objective (e.g., h... | 2024.emnlp-main.85 | 10.18653/v1/2024.emnlp-main.85 | null | 2402.19085 | title_snapshot | [
0.00151797104626894,
0.014752042479813099,
-0.011427712626755238,
0.03481510654091835,
0.02671651914715767,
0.03655216097831726,
0.023906482383608818,
0.007412314880639315,
-0.001677082385867834,
-0.06700368225574493,
-0.008547592908143997,
0.026218166574835777,
-0.08887369930744171,
-0.03... |
Mitigating Matthew Effect: Multi-Hypergraph Boosted Multi-Interest Self-Supervised Learning for Conversational Recommendation | https://aclanthology.org/2024.emnlp-main.86/ | [
"Yongsen Zheng",
"Ruilin Xu",
"Guohua Wang",
"Liang Lin",
"Kwok-Yan Lam"
] | The Matthew effect is a big challenge in Recommender Systems (RSs), where popular items tend to receive increasing attention, while less popular ones are often overlooked, perpetuating existing disparities. Although many existing methods attempt to mitigate Matthew effect in the static or quasi-static recommendation sc... | 2024.emnlp-main.86 | 10.18653/v1/2024.emnlp-main.86 | null | null | null | [
0.04705854877829552,
0.009785507805645466,
0.01649535819888115,
0.04786214977502823,
0.03932345658540726,
-0.015498715452849865,
0.03110615350306034,
0.009193030185997486,
-0.011355043388903141,
-0.04683712497353554,
-0.015906479209661484,
0.007616645190864801,
-0.0659673735499382,
-0.0090... |
Advancing Event Causality Identification via Heuristic Semantic Dependency Inquiry Network | https://aclanthology.org/2024.emnlp-main.87/ | [
"Haoran Li",
"Qiang Gao",
"Hongmei Wu",
"Li Huang"
] | Event Causality Identification (ECI) focuses on extracting causal relations between events in texts. Existing methods for ECI primarily rely on causal features and external knowledge. However, these approaches fall short in two dimensions: (1) causal features between events in a text often lack explicit clues, and (2) ... | 2024.emnlp-main.87 | 10.18653/v1/2024.emnlp-main.87 | null | 2409.13621 | title_snapshot | [
0.0009623495861887932,
-0.01697709783911705,
-0.03680754452943802,
0.04813896864652634,
0.039065223187208176,
0.016300328075885773,
0.033820293843746185,
-0.0036107581108808517,
0.009229844436049461,
0.0007226909510791302,
0.0002272586280014366,
0.035730719566345215,
-0.03326277807354927,
... |
Exploring Union and Intersection of Visual Regions for Generating Questions, Answers, and Distractors | https://aclanthology.org/2024.emnlp-main.88/ | [
"Wenjian Ding",
"Yao Zhang",
"Jun Wang",
"Adam Jatowt",
"Zhenglu Yang"
] | Multiple-choice visual question answering (VQA) is to automatically choose a correct answer from a set of choices after reading an image. Existing efforts have been devoted to a separate generation of an image-related question, a correct answer, or challenge distractors. By contrast, we turn to a holistic generation an... | 2024.emnlp-main.88 | 10.18653/v1/2024.emnlp-main.88 | null | null | null | [
0.04653206840157509,
-0.005059499759227037,
0.008170981891453266,
0.07697029411792755,
0.03710420802235603,
0.033467650413513184,
0.008230099454522133,
0.02348836325109005,
-0.06162790209054947,
-0.016412535682320595,
-0.07173298299312592,
0.021368330344557762,
-0.08158183097839355,
-0.026... |
UniFashion: A Unified Vision-Language Model for Multimodal Fashion Retrieval and Generation | https://aclanthology.org/2024.emnlp-main.89/ | [
"Xiangyu Zhao",
"Yuehan Zhang",
"Wenlong Zhang",
"Xiao-Ming Wu"
] | The fashion domain encompasses a variety of real-world multimodal tasks, including multimodal retrieval and multimodal generation. The rapid advancements in artificial intelligence generated content, particularly in technologies like large language models for text generation and diffusion models for visual generation, ... | 2024.emnlp-main.89 | 10.18653/v1/2024.emnlp-main.89 | null | 2408.11305 | title_snapshot | [
0.008137303404510021,
-0.04083016887307167,
-0.020412079989910126,
0.03544800356030464,
0.059127889573574066,
0.008187918923795223,
0.005924435332417488,
0.029489265754818916,
0.003984777722507715,
-0.03655628859996796,
-0.055629879236221313,
0.008737031370401382,
-0.07316635549068451,
-0.... |
Tracking the perspectives of interacting language models | https://aclanthology.org/2024.emnlp-main.90/ | [
"Hayden Helm",
"Brandon Duderstadt",
"Youngser Park",
"Carey Priebe"
] | Large language models (LLMs) are capable of producing high quality information at unprecedented rates. As these models continue to entrench themselves in society, the content they produce will become increasingly pervasive in databases that are, in turn, incorporated into the pre-training data, fine-tuning data, retrie... | 2024.emnlp-main.90 | 10.18653/v1/2024.emnlp-main.90 | null | 2406.11938 | title_snapshot | [
-0.036542948335409164,
-0.009658931754529476,
-0.00946453120559454,
0.01423631515353918,
0.05004805326461792,
-0.0029457728378474712,
0.013694853521883488,
0.05063813552260399,
-0.042018819600343704,
0.009560099802911282,
-0.025574099272489548,
0.03531487286090851,
-0.055713195353746414,
-... |
MAR: Matching-Augmented Reasoning for Enhancing Visual-based Entity Question Answering | https://aclanthology.org/2024.emnlp-main.91/ | [
"Zhengxuan Zhang",
"Yin Wu",
"Yuyu Luo",
"Nan Tang"
] | A multimodal large language model MLLMs may struggle with answering visual-based (personal) entity questions (VEQA), such as ”who is A?” or ”who is A that B is talking to?” for various reasons, e.g., the absence of the name of A in the caption or the inability of MLLMs to recognize A, particularly for less common entit... | 2024.emnlp-main.91 | 10.18653/v1/2024.emnlp-main.91 | null | null | null | [
0.011737590655684471,
0.0018006153404712677,
0.020066602155566216,
0.03846284747123718,
0.041658591479063034,
0.01738496869802475,
0.015403186902403831,
0.007586092222481966,
-0.05004303529858589,
-0.0015497143613174558,
-0.05740150064229965,
0.04383625090122223,
-0.07048533111810684,
-0.0... |
Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones? | https://aclanthology.org/2024.emnlp-main.92/ | [
"Zhe Yang",
"Yichang Zhang",
"Tianyu Liu",
"Jian Yang",
"Junyang Lin",
"Chang Zhou",
"Zhifang Sui"
] | Large language models (LLMs) have demonstrated impressive capabilities, but still suffer from inconsistency issues (e.g. LLMs can react differently to disturbances like rephrasing or inconsequential order change). In addition to these inconsistencies, we also observe that LLMs, while capable of solving hard problems, c... | 2024.emnlp-main.92 | 10.18653/v1/2024.emnlp-main.92 | null | 2406.12809 | title_snapshot | [
-0.018833892419934273,
-0.033283766359090805,
-0.040846604853868484,
0.05861910432577133,
0.028179645538330078,
0.021250421181321144,
0.032275471836328506,
0.03539959713816643,
-0.030079368501901627,
-0.006744174286723137,
-0.025665810331702232,
0.04202497377991676,
-0.06986397504806519,
-... |
Watch Every Step! LLM Agent Learning via Iterative Step-level Process Refinement | https://aclanthology.org/2024.emnlp-main.93/ | [
"Weimin Xiong",
"Yifan Song",
"Xiutian Zhao",
"Wenhao Wu",
"Xun Wang",
"Ke Wang",
"Cheng Li",
"Wei Peng",
"Sujian Li"
] | Large language model agents have exhibited exceptional performance across a range of complex interactive tasks. Recent approaches have utilized tuning with expert trajectories to enhance agent performance, yet they primarily concentrate on outcome rewards, which may lead to errors or suboptimal actions due to the absen... | 2024.emnlp-main.93 | 10.18653/v1/2024.emnlp-main.93 | null | 2406.11176 | title_snapshot | [
-0.014433405362069607,
-0.014537408947944641,
0.013237657956779003,
0.03118179179728031,
0.03887549042701721,
-0.018022051081061363,
0.05798972025513649,
0.013119824230670929,
0.003295305883511901,
-0.009043410420417786,
-0.02930869162082672,
0.021171946078538895,
-0.04945008084177971,
-0.... |
Standardize: Aligning Language Models with Expert-Defined Standards for Content Generation | https://aclanthology.org/2024.emnlp-main.94/ | [
"Joseph Marvin Imperial",
"Gail Forey",
"Harish Tayyar Madabushi"
] | Domain experts across engineering, healthcare, and education follow strict standards for producing quality content such as technical manuals, medication instructions, and children’s reading materials. However, current works in controllable text generation have yet to explore using these standards as references for cont... | 2024.emnlp-main.94 | 10.18653/v1/2024.emnlp-main.94 | null | 2402.12593 | title_snapshot | [
0.00015081267338246107,
0.009730806574225426,
-0.017666956409811974,
0.022651294246315956,
0.07149842381477356,
-0.028939343988895416,
0.04097931832075119,
0.03510782867670059,
-0.021681975573301315,
-0.003224526299163699,
-0.03151734918355942,
0.0238321665674448,
-0.05533638224005699,
-0.... |
Cross-domain NER with Generated Task-Oriented Knowledge: An Empirical Study from Information Density Perspective | https://aclanthology.org/2024.emnlp-main.95/ | [
"Zhihao Zhang",
"Sophia Yat Mei Lee",
"Junshuang Wu",
"Dong Zhang",
"Shoushan Li",
"Erik Cambria",
"Guodong Zhou"
] | Cross-domain Named Entity Recognition (CDNER) is crucial for Knowledge Graph (KG) construction and natural language processing (NLP), enabling learning from source to target domains with limited data. Previous studies often rely on manually collected entity-relevant sentences from the web or attempt to bridge the gap b... | 2024.emnlp-main.95 | 10.18653/v1/2024.emnlp-main.95 | null | null | null | [
-0.02115800976753235,
-0.02594301663339138,
0.007851000875234604,
0.04901689291000366,
0.01952718198299408,
-0.010803936049342155,
0.026238173246383667,
-0.006482367403805256,
0.010857880115509033,
-0.007787251845002174,
-0.03666304796934128,
0.02557472698390484,
-0.04365144297480583,
-0.0... |
Glue pizza and eat rocks - Exploiting Vulnerabilities in Retrieval-Augmented Generative Models | https://aclanthology.org/2024.emnlp-main.96/ | [
"Zhen Tan",
"Chengshuai Zhao",
"Raha Moraffah",
"Yifan Li",
"Song Wang",
"Jundong Li",
"Tianlong Chen",
"Huan Liu"
] | Retrieval-Augmented Generative (RAG) models enhance Large Language Models (LLMs) by integrating external knowledge bases, improving their performance in applications like fact-checking and information searching. In this paper, we demonstrate a security threat where adversaries can exploit the openness of these knowledg... | 2024.emnlp-main.96 | 10.18653/v1/2024.emnlp-main.96 | null | 2406.19417 | title_snapshot | [
0.0011042025871574879,
-0.023920709267258644,
-0.03062412515282631,
0.07297321408987045,
0.04933050647377968,
-0.017827169969677925,
0.03234107047319412,
-0.0014821034856140614,
-0.02242937497794628,
-0.0069193062372505665,
-0.0740235298871994,
0.04534187167882919,
-0.07537799328565598,
0.... |
Predicate Debiasing in Vision-Language Models Integration for Scene Graph Generation Enhancement | https://aclanthology.org/2024.emnlp-main.97/ | [
"Yuxuan Wang",
"Xiaoyuan Liu"
] | Scene Graph Generation (SGG) provides basic language representation of visual scenes, requiring models to grasp complex and diverse semantics between objects. This complexity and diversity in SGG leads to underrepresentation, where parts of triplet labels are rare or even unseen during training, resulting in imprecise ... | 2024.emnlp-main.97 | 10.18653/v1/2024.emnlp-main.97 | null | 2403.16184 | title_snapshot | [
0.0020087112206965685,
0.002705978462472558,
0.021774331107735634,
0.04826192557811737,
0.02358902618288994,
0.011878290213644505,
0.0328608974814415,
0.025581208989024162,
-0.03177270293235779,
-0.04025845602154732,
-0.016091901808977127,
0.0064426567405462265,
-0.10040069371461868,
-0.00... |
SHIELD: Evaluation and Defense Strategies for Copyright Compliance in LLM Text Generation | https://aclanthology.org/2024.emnlp-main.98/ | [
"Xiaoze Liu",
"Ting Sun",
"Tianyang Xu",
"Feijie Wu",
"Cunxiang Wang",
"Xiaoqian Wang",
"Jing Gao"
] | Large Language Models (LLMs) have transformed machine learning but raised significant legal concerns due to their potential to produce text that infringes on copyrights, resulting in several high-profile lawsuits. The legal landscape is struggling to keep pace with these rapid advancements, with ongoing debates about w... | 2024.emnlp-main.98 | 10.18653/v1/2024.emnlp-main.98 | null | 2406.12975 | title_snapshot | [
-0.005137959495186806,
-0.021561071276664734,
-0.03347970172762871,
0.04855528473854065,
0.06880779564380646,
-0.03188662976026535,
0.02633793093264103,
-0.001164209796115756,
-0.02093128114938736,
0.00573299266397953,
-0.02944810315966606,
0.030030397698283195,
-0.05480257794260979,
0.010... |
MatchTime: Towards Automatic Soccer Game Commentary Generation | https://aclanthology.org/2024.emnlp-main.99/ | [
"Jiayuan Rao",
"Haoning Wu",
"Chang Liu",
"Yanfeng Wang",
"Weidi Xie"
] | Soccer is a globally popular sport with a vast audience, in this paper, we consider constructing an automatic soccer game commentary model to improve the audiences’ viewing experience. In general, we make the following contributions: *First*, observing the prevalent video-text misalignment in existing datasets, we manu... | 2024.emnlp-main.99 | 10.18653/v1/2024.emnlp-main.99 | null | 2406.18530 | title_snapshot | [
0.002456423593685031,
-0.03168235346674919,
0.0010926872491836548,
0.01994442753493786,
0.012249995023012161,
-0.017725428566336632,
0.029361462220549583,
0.05727297067642212,
0.0044654798693954945,
-0.02386181429028511,
0.003846193663775921,
0.031717587262392044,
-0.0796392485499382,
-0.0... |
Rethinking Token Reduction for State Space Models | https://aclanthology.org/2024.emnlp-main.100/ | [
"Zheng Zhan",
"Yushu Wu",
"Zhenglun Kong",
"Changdi Yang",
"Yifan Gong",
"Xuan Shen",
"Xue Lin",
"Pu Zhao",
"Yanzhi Wang"
] | Recent advancements in State Space Models (SSMs) have attracted significant interest, particularly in models optimized for parallel training and handling long-range dependencies. Architectures like Mamba have scaled to billions of parameters with selective SSM. To facilitate broader applications using Mamba, exploring ... | 2024.emnlp-main.100 | 10.18653/v1/2024.emnlp-main.100 | null | 2410.14725 | title_snapshot | [
-0.039952073246240616,
-0.03499869629740715,
-0.01400406751781702,
0.04674512892961502,
0.0422869473695755,
0.044382717460393906,
0.01915176585316658,
0.004826775752007961,
-0.03109630197286606,
-0.026699086651206017,
0.01662161387503147,
-0.005542336963117123,
-0.07490966469049454,
-0.006... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.