arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

Annual Meeting of the Association for Computational Linguistics · 会议 · Natural Language Processing

共收录 10305
2506.02998 2026-02-03 cs.CL

A Multi-Agent Framework for Mitigating Dialect Biases in Privacy Policy Question-Answering Systems

一种缓解隐私政策问答系统方言偏见的多智能体框架

Đorđe Klisura, Astrid R Bernaga Torres, Anna Karen Gárate-Escamilla, Rajesh Roshan Biswal, Ke Yang, Hilal Pataci, Anthony Rios

机构 * University of Texas at San Antonio(德克萨斯大学圣安东尼奥分校) Tecnológico de Monterrey(蒙特雷技术学院)

AI总结 本文提出一种多智能体框架,通过方言智能体和隐私政策智能体协作,提升隐私政策问答系统对非标准方言的处理能力,有效缓解方言偏见问题。

Comments Accepted to ACL 2025 Main Conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2403.00068 2026-02-03 cs.CV

Towards Artwork Explanation in Large-scale Vision Language Models

迈向大规模视觉语言模型中的艺术作品解释

Kazuki Hayashi, Yusuke Sakai, Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

机构 * Nara Institute of Science and Technology(奈良科学技术大学) The University of Tokyo(东京大学)

AI总结 本文提出艺术作品解释生成任务,评估大规模视觉语言模型在整合语言和视觉信息以及从图像中获取知识的能力。

Comments Accepted to ACL 2024 (Main Conference)

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.00038 2026-02-03 cs.CY cs.AI

LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion

LSSF: 通过低秩安全子空间融合实现大语言模型的安全对齐

Guanghao Zhou, Panjia Qiu, Cen Chen, Hongyu Li, Mingyuan Chu, Xin Zhang, Jun Zhou

机构 * East China Normal University(华东师范大学) Ant Group(蚂蚁集团)

AI总结 LSSF通过低秩安全子空间融合技术,有效恢复微调大语言模型的安全对齐,同时对性能影响较小。

Comments Accepted in ACL 2025 Main Conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.22439 2026-02-02 cs.CL

Stop Jostling: Adaptive Negative Sampling Reduces the Marginalization of Low-Resource Language Tokens by Cross-Entropy Loss

停止推搡:自适应负采样减少交叉熵损失对低资源语言标记的边缘化

Galim Turumtaev

AI总结 本文提出自适应负采样技术,通过减少交叉熵损失对低资源语言标记的边缘化影响,提升模型对低资源语言的表示能力。

Comments Accepted at LoResLM 2025 (COLING 2025 workshop). Oral presentation

Journal ref In Proceedings of the First Workshop on Language Models for Low-Resource Languages (LoResLM 2025), pages 373-386, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2106.10199 2026-01-30 cs.LG cs.CL

BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models

BitFit: 一种基于Transformer的掩码语言模型的简单参数高效微调方法

Elad Ben-Zaken, Shauli Ravfogel, Yoav Goldberg

AI总结 BitFit是一种针对Transformer掩码语言模型的简单参数高效微调方法,通过仅修改模型的偏置项来实现与完整微调相当甚至更优的性能。

Comments Accepted at ACL 2022 main conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2501.01203 2026-01-30 cs.SI

HetGCoT: Heterogeneous Graph-Enhanced Chain-of-Thought LLM Reasoning for Academic Question Answering

HetGCoT:异构图增强的链式思考LLM推理用于学术问答

Runsong Jia, Mengjia Wu, Ying Ding, Jie Lu, Yi Zhang

AI总结 HetGCoT通过异构图增强链式思考LLM推理,提升学术问答的可解释性和准确性。

Comments Findings of the Association for Computational Linguistics: EMNLP 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2406.00197 2026-01-30 cs.CL

Re3: A Holistic Framework and Dataset for Modeling Collaborative Document Revision

Re3: 一个用于协作文档修订建模的综合框架和数据集

Qian Ruan, Ilia Kuznetsov, Iryna Gurevych

机构 * Ubiquitous Knowledge Processing Lab (UKP Lab)(通用知识处理实验室) Department of Computer Science(计算机科学系) Hessian Center for AI (hessian.AI)(黑森人工智能中心)

AI总结 Re3通过构建学术领域协作文档修订数据集,探索协作修订机制并评估LLM在自动化编辑分析与协作中的能力。

Comments accepted to ACL2024 main

Journal ref Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics ACL 2024 (Volume 1: Long Papers)

详情

展开后加载摘要…

URL PDF HTML 收藏
2203.09629 2026-01-30 cs.CL

HiStruct+: Improving Extractive Text Summarization with Hierarchical Structure Information

HiStruct+: 通过层次结构信息提升基于提取的文本摘要

Qian Ruan, Malte Ostendorff, Georg Rehm

AI总结 HiStruct+通过显式注入层次结构信息,提升提取式摘要的ROUGEs指标,在PubMed和arXiv数据集上取得显著效果。

Comments 17 pages, 3 figures, to be published in Findings ACL 2022

Journal ref Findings of the Association for Computational Linguistics ACL 2022

详情

展开后加载摘要…

URL PDF HTML 收藏
2408.04628 2026-01-29 cs.CL cs.AI cs.CV

LogogramNLP: Comparing Visual and Textual Representations of Ancient Logographic Writing Systems for NLP

LogogramNLP: 比较古代象形文字系统的视觉与文本表示在自然语言处理中的应用

Danlu Chen, Freda Shi, Aditi Agarwal, Jacobo Myerston, Taylor Berg-Kirkpatrick

机构 * UC San Diego(UC圣地亚哥大学) University of Waterloo(滑铁卢大学)

AI总结 LogogramNLP通过比较视觉与文本表示,为古代象形文字的NLP分析提供首个基准,展示视觉方法在部分任务中的优势。

Comments correct wrong refs, typos

Journal ref ACL 2024, long paper

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.20465 2026-01-29 cs.CL

BMAM: Brain-inspired Multi-Agent Memory Framework

基于大脑的多智能体记忆框架 BMAM

Yang Li, Jiaxiang Liu, Yusong Wang, Yujie Wu, Mingkun Xu

机构 * Guangdong Institute of Intelligence Science and Technology(广东智能科学与技术研究院) Institute of Science Tokyo(东京科学研究所) The Hong Kong Polytechnic University(香港理工大学)

AI总结 BMAM 提出了一种基于大脑认知机制的多智能体记忆框架,通过分解记忆为事件型、语义型等子系统,提升长时间交互中的信息保持和行为一致性,实验显示其在 LoCoMo 基准上达到 78.45% 的准确率。

Comments Submitted to ACL (ARR 2026 January submission); non-anonymous preprint

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.16656 2026-01-29 cs.AI

NUMINA: A Natural Understanding Benchmark for Multi-dimensional Intelligence and Numerical Reasoning Abilities

NUMINA:多维智能与数值推理能力的自然理解基准

Changyu Zeng, Yifan Wang, Zimu Wang, Wei Wang, Zhengni Yang, Muyi Bao, Jiming Xiao, Anh Nguyen, Yutao Yue

机构 * School of Advanced Technology, Xi’an Jiaotong-Liverpool University(西安交通大学利物浦大学先进技术学院) Department of Computer Science, University of Liverpool(利物浦大学计算机科学系) The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州)) Institute of Deep Perception Technology, JITRI(深度感知技术研究所)

AI总结 NUMINA是一个用于多维智能和数值推理能力的自然理解基准,通过多尺度注释和自动化注释流程提升多模态室内感知理解,揭示当前LLM在三维空间计算中的不足。

Journal ref Findings of the Association for Computational Linguistics: EMNLP 2025, pages 22575--22590

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.19447 2026-01-28 cs.CL cs.AI

KG-CRAFT: Knowledge Graph-based Contrastive Reasoning with LLMs for Enhancing Automated Fact-checking

基于知识图谱的对比推理:利用大语言模型增强自动事实核查

Vítor N. Lourenço, Aline Paes, Tillman Weyde, Audrey Depeige, Mohnish Dubey

机构 * Universidade Federal Fluminense(联邦Fluminense大学) Amazon(亚马逊) City St George’s, University of London(伦敦大学圣乔治学院)

AI总结 KG-CRAFT通过结合知识图谱和大语言模型,提升自动事实核查的准确性与性能。

Comments Accepted to publication at the 19th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.19096 2026-01-28 cs.CL

PsyProbe: Proactive and Interpretable Dialogue through User State Modeling for Exploratory Counseling

PsyProbe: 通过用户状态建模实现探索性辅导的主动与可解释对话

Sohhyung Park, Hyunji Kang, Sungzoon Cho, Dongil Kim

机构 * Department of Industrial Engineering, Seoul National University(工业工程系,首尔国立大学) Department of Education, Seoul National University(教育系,首尔国立大学)

AI总结 PsyProbe通过系统性用户状态建模和主动提问提升探索性辅导的互动效果和专业性。

Comments In Findings of the Association for Computational Linguistics: EACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.18527 2026-01-27 cs.CL

Exploring Fine-Tuning for In-Context Retrieval and Efficient KV-Caching in Long-Context Language Models

探索针对长上下文语言模型的微调以实现上下文检索和高效的KV缓存

Francesco Maria Molfese, Momchil Hardalov, Rexhina Blloshmi, Bill Byrne, Adrià de Gispert

机构 * Sapienza University of Rome(罗马萨皮恩扎大学) Amazon AGI(亚马逊人工智能研究院)

AI总结 本文研究了长上下文语言模型在微调策略下的性能提升及KV缓存压缩下的鲁棒性,展示了领域内和跨领域任务中的不同表现。

Comments European Chapter of the Association for Computational Linguistics EACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.18415 2026-01-27 cs.CL cs.SD eess.AS

Pisets: A Robust Speech Recognition System for Lectures and Interviews

Pisets:一种用于讲座和访谈的鲁棒语音识别系统

Ivan Bondarenko, Daniil Grebenkin, Oleg Sedukhin, Mikhail Klementev, Roman Derunets, Lyudmila Budneva

机构 * Novosibirsk State University(新西伯利亚州立大学) Siberian Neuronets LLC(西伯利亚神经网络有限公司)

AI总结 Pisets通过三组件架构和课程学习提升语音识别鲁棒性,适用于讲座和访谈场景。

Journal ref Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 3: Industry Track), pp. 988-997

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.13551 2026-01-27 cs.AI

Tandem Training for Language Models

语言模型的串联训练

Robert West, Ashton Anderson, Ece Kamar, Eric Horvitz

机构 * EPFL(瑞士联邦理工学院) University of Toronto(多伦多大学) Microsoft(微软公司)

AI总结 本文提出串联训练方法,通过强化学习促进语言模型在任务中保持可解释性,使模型能适应较弱协作者并保持高准确性。

Journal ref Proceedings of the 2026 Conference of the European Chapter of the Association for Computational Linguistics (EACL)

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.17203 2026-01-27 cs.CL

Relating Word Embedding Gender Biases to Gender Gaps: A Cross-Cultural Analysis

将词嵌入性别偏见与性别差距联系起来:跨文化分析

Scott Friedman, Sonja Schmer-Galunder, Anthony Chen, Jeffrey Rye

机构 * SIFT

AI总结 本文通过量化词嵌入中的性别偏见,分析不同文化背景下教育、政治、经济和健康领域的性别差距。

Comments 7 pages, 5 figures. Presented at the First Workshop on Gender Bias in Natural Language Processing (GeBNLP 2019)

Journal ref In Proceedings of the First Workshop on Gender Bias in Natural Language Processing, pages 18-24, Florence, Italy. Association for Computational Linguistics (2019)

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.22590 2026-01-27 cs.AI cs.CL cs.IR

ATOM: AdapTive and OptiMized dynamic temporal knowledge graph construction using LLMs

ATOM: 基于LLMs的自适应与优化动态时间知识图谱构建

Yassir Lairgi, Ludovic Moncla, Khalid Benabdeslem, Rémy Cazabet, Pierre Cléau

机构 * INSA Lyon, CNRS, UCBL, LIRIS, UMR5205(里里斯研究所) GAUC, Lyon, France(里昂高研院)

AI总结 ATOM通过原子事实提取和双时间建模,实现动态时间知识图谱的高完整性和稳定性,提升实时数据处理能力。

Comments Accepted at the Findings of the Association for Computational Linguistics: EACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.04945 2026-01-27 cs.CL cs.AI cs.CV

Towards Robust Evaluation of Visual Activity Recognition: Resolving Verb Ambiguity with Sense Clustering

面向视觉活动识别的稳健评估:通过语义聚类解决动词歧义

Louie Hong Yao, Nicholas Jarvis, Tianyu Jiang

机构 * University of Cincinnati(辛辛那提大学)

AI总结 本文提出基于语义聚类的视觉活动识别评估方法,通过解决动词歧义问题提升评估的鲁棒性。

Comments Accepted to Findings of the Association for Computational Linguistics: EACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.08660 2026-01-27 cs.CL cs.LG

The Impact of Automatic Speech Transcription on Speaker Attribution

自动语音转录对说话人归因的影响

Cristina Aggazzotti, Matthew Wiesner, Elizabeth Allyn Smith, Nicholas Andrews

机构 * Johns Hopkins University(约翰霍普金斯大学) Université du Québec à Montréal(魁北克大学蒙特利尔分校)

AI总结 本文研究了自动语音转录对说话人归因性能的影响,发现即使在存在转录错误的情况下,归因性能仍保持良好,可能是因为ASR转录错误能揭示说话人身份特征。

Comments latest version added TACL journal DOI to metadata and a missing citation

Journal ref Transactions of the Association for Computational Linguistics (2025) 13: 1578-1596

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.04772 2026-01-26 cs.CL

Identifying Reliable Evaluation Metrics for Scientific Text Revision

识别科学文本修订中的可靠评估指标

Léane Jourdan, Florian Boudin, Richard Dufour, Nicolas Hernandez

机构 * Nantes Université, École Centrale Nantes, CNRS, LS2N, UMR 6004(南特大学、中央理工学院、CNRS、LS2N、UMR 6004)

AI总结 本文提出通过结合LLM评估与领域特定指标的方法,以更准确评估科学文本修订的质量。

Comments V5 contains the English version, (ACL 2025 main, 26 pages) and V4 contains the French version (TALN 2025, 32 pages), both with corrected results for cramer's v and pairwise accuracy

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.15809 2026-01-23 cs.CL

SteerEval: Inference-time Interventions Strengthen Multilingual Generalization in Neural Summarization Metrics

SteerEval: 在推理时间干预增强神经摘要度量的多语言泛化

Silvia Casola, Ryan Soh-Eun Shim, Felicia Körner, Yuchen Mao, Barbara Plank

机构 * MaiNLP, Center for Information and Language Processing, LMU Munich(MaiNLP、信息与语言处理中心、慕尼黑大学) Language Science and Technology, Saarland University(语言科学与技术、萨尔兰大学)

AI总结 SteerEval通过在推理时干预激活向英语基准倾斜,提升多语言神经摘要度量的泛化能力。

Comments Submitted to ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.15487 2026-01-23 cs.AI cs.CL cs.MA

MiRAGE: A Multiagent Framework for Generating Multimodal Multihop Question-Answer Dataset for RAG Evaluation

MiRAGE:一种多智能体框架,用于生成多模态多跳问题-答案数据集以评估RAG系统

Chandan Kumar Sahu, Premith Kumar Chilukuri, Matthew Hetrich

机构 * ABB Inc(ABB公司)

AI总结 MiRAGE通过多智能体框架生成多模态多跳问题-答案数据集,提升RAG系统评估的准确性和复杂性。

Comments 12 pages, 2 figures, Submitted to ACL

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.15337 2026-01-23 cs.LG cs.CL

Language Models Entangle Language and Culture

语言模型融合语言与文化

Shourya Jain, Paras Chopra

机构 * Lossfunk

AI总结 研究发现语言选择影响LLM生成答案的文化上下文,导致低资源语言回答质量较低。

Comments Accepted at LM4UC Workshop at AAAI'26, Submitted to ACL 2026. 17 pages, 7 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12471 2026-01-23 cs.CL cs.AI

Knowing When to Abstain: Medical LLMs Under Clinical Uncertainty

知何时退避:医疗大语言模型在临床不确定性中的表现

Sravanthi Machcha, Sushrita Yerra, Sahil Gupta, Aishwarya Sahoo, Sharmin Sultana, Hong Yu, Zonghai Yao

机构 * Manning College of Information and Computer Sciences, UMass Amherst, MA, USA(马萨诸塞大学阿姆赫斯特曼宁信息与计算机科学学院) Center for Healthcare Organization and Implementation Research, VA Bedford Health Care(医疗组织与实施研究中心) Miner School of Computer and Information Sciences, UMass Lowell, MA, USA(米纳尔计算机与信息科学学院)

AI总结 本文提出MedAbstain基准,探讨医疗LLM在临床不确定性中的退避能力,发现显式退避选项能显著提升安全性,而模型规模和提示方法效果有限。

Comments Equal contribution for the first two authors; To appear in proceedings of the Main Conference of the European Chapter of the Association for Computational Linguistics (EACL) 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12061 2026-01-23 cs.CL cs.AI

Codebook-Injected Dialogue Segmentation for Multi-Utterance Constructs Annotation: LLM-Assisted and Gold-Label-Free Evaluation

用于多轮对话结构标注的代码表注入对话分割:LLM辅助且无需黄金标签的评估

Jinsook Lee, Kirk Vanacore, Zhuqian Zhou, Bakhtawar Ahtisham, Jeanine Grutter, Rene F. Kizilcec

机构 * Cornell University(康奈尔大学) LMU Muinich(慕尼黑大学)

AI总结 本文提出一种基于LLM的对话分割方法,通过代码表注入提升分割一致性,并在无黄金标签情况下评估不同分割器的性能,发现需根据下游任务优化分割策略。

Comments Under Review for ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.23761 2026-01-23 cs.SE cs.AI cs.MA

TDFlow: Agentic Workflows for Test Driven Development

TDFlow: 为测试驱动开发设计的代理工作流

Kevin Han, Siddharth Maddikayala, Tim Knappe, Om Patel, Austen Liao, Amir Barati Farimani

机构 * Carnegie Mellon University(卡内基梅隆大学) UC San Diego(南加州大学) Johns Hopkins University(约翰霍普金斯大学)

AI总结 TDFlow通过测试驱动的工作流实现人类水平的测试解析,提升软件修复性能。

Comments Published in the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026 Main Conference)

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.10695 2026-01-23 cs.LG cs.AI cs.CL

Introducing Verification Task of Set Consistency with Set-Consistency Energy Networks

引入集合一致性验证任务与集合一致性能量网络

Mooho Song, Hyeryung Son, Jay-Yoon Lee

机构 * Seoul National University(首尔国立大学)

AI总结 本文提出集合一致性验证任务及SC-Energy模型,通过对比损失框架提升多陈述逻辑一致性验证性能,并发布新数据集

Journal ref Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), Long Papers

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.08134 2026-01-22 cs.CL

How Reliable are Confidence Estimators for Large Reasoning Models? A Systematic Benchmark on High-Stakes Domains

大型推理模型的置信度估计有多可靠?对高风险领域的系统基准测试

Reza Khanmohammadi, Erfan Miahi, Simerjot Kaur, Ivan Brugere, Charese H. Smiley, Kundan Thind, Mohammad M. Ghassemi

AI总结 本文通过系统基准测试,评估了大型推理模型置信度估计方法的可靠性,发现基于文本的编码器在歧视方面表现最佳,而结构感知模型在校准方面表现最佳,揭示了当前方法的局限性。

Comments Accepted to the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026) main conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.18085 2026-01-22 cs.CL cs.AI cs.LG

Temporal Relation Extraction in Clinical Texts: A Span-based Graph Transformer Approach

临床文本中的时间关系抽取:一种基于跨度的图变换器方法

Rochana Chaturvedi, Peyman Baghershahi, Sourav Medya, Barbara Di Eugenio

AI总结 本文提出GRAPHTREX方法,通过结合基于跨度的实体关系抽取、临床预训练语言模型和异构图变换器,提升临床文本中时间关系抽取的准确率和长距离关系识别能力。

Comments Introducing a novel method for joint extraction of medical events and temporal relations from free-text, leveraging clinical LPLMs and Heterogeneous Graph Transformers, achieving a 5.5% improvement over the previous state-of-the-art and up to 8.9% on long-range relations

Journal ref Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏