Code-Switching Curriculum Learning for Multilingual Transfer in LLMs
Comments To appear in Findings of ACL 2025
期刊&会议
ACL
Annual Meeting of the Association for Computational Linguistics · 会议 · Natural Language Processing
Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-thoughts Critic
机构 * Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所信息处理实验室） ； University of Chinese Academy of Sciences（中国科学院大学） ； Xiaohongshu Inc（小红书公司）
Comments Accepted at ACL 2025 Findings
Code-Switching Red-Teaming: LLM Evaluation for Safety and Multilingual Understanding
Comments To appear in ACL 2025
MTVQA: Benchmarking Multilingual Text-Centric Visual Question Answering
机构 * ByteDance Inc.（字节跳动公司） ； Huazhong University of Science and Technology（华中科技大学） ； University of Minnesota（明尼苏达大学） ； Cornell University（康奈尔大学）
Comments Accepted by ACL 2025 findings
DREsS: Dataset for Rubric-based Essay Scoring on EFL Writing
Comments To appear in ACL 2025. arXiv admin note: text overlap with arXiv:2310.05191
Dialect Normalization using Large Language Models and Morphological Rules
机构 * Archimedes, Athena Research Center, Greece（阿基米德，雅典娜研究中心，希腊） ； Department of Informatics and Telecommunications, NKUA（信息与电信系，NKUA） ； Department of Informatics, Athens University of Economics and Business, Greece（信息系，雅典经济与商业大学，希腊） ； Department of Computer Science, George Mason University（计算机科学系，乔治·梅森大学）
Comments 19 pages, 18 figures, to be published in the Findings of the Association for Computational Linguistics 2025
Educators' Perceptions of Large Language Models as Tutors: Comparing Human and AI Tutors in a Blind Text-only Setting
Comments Accepted to BEA@ACL 2025
VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism
机构 * School of Computer Science and Engineering, Sun Yat-Sen University（中山大学计算机科学与工程学院） ； Guangdong Laboratory of Artificial Intelligence and Digital Economy（广东省人工智能与数字经济实验室） ； Huawei Technologies Co., Ltd（华为技术有限公司）
Comments Accepted by ACL 2025 main
TableDreamer: Progressive and Weakness-guided Data Synthesis from Scratch for Table Instruction Tuning
机构 * Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所） ； School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院） ； Baidu Inc（百度公司）
Comments 27 pages, 19 figures, Findings of ACL 2025
CoMuMDR: Code-mixed Multi-modal Multi-domain corpus for Discourse paRsing in conversations
机构 * Indian Institute of Technology Kanpur (IIT Kanpur)（印度理工学院坎浦尔学院）
Comments Accepted at ACL Findings 2025 (16 pages: 5 pages main content + 3 pages references + 8 pages appendix)
EtiCor++: Towards Understanding Etiquettical Bias in LLMs
机构 * Indian Institute of Technology Kanpur (IIT Kanpur)（印度理工学院坎浦尔学院）
Comments Accepted at ACL Findings 2025, 22 pages (9 pages main content + 4 pages references + 9 pages appendix)
Silencing Empowerment, Allowing Bigotry: Auditing the Moderation of Hate Speech on Twitch
机构 * Indian Institute of Science（印度科学研究院） ； University of Chicago（芝加哥大学） ； Indian Institute of Technology, Bombay（班加罗尔印度理工学院）
Comments Accepted to ACL 2025 (main) conference
Mutual-Taught for Co-adapting Policy and Reward Models
机构 * School of Computer Science and Engineering, Sun Yat-sen University, China（计算机科学与工程学院，中山大学） ； Alibaba Group（阿里巴巴集团）
Comments Accepted to ACL 2025 (Main Conference)
Optimized Text Embedding Models and Benchmarks for Amharic Passage Retrieval
机构 * University of Amsterdam（阿姆斯特丹大学）
Comments 10 pages (excl. refs/appendix), 10 figures. Accepted to ACL 2025 Findings. Kidist and Yosef contributed equally to this work. Public resources: https://github.com/kidist-amde/amharic-ir-benchmarks
Mixture of Decoding: An Attention-Inspired Adaptive Decoding Strategy to Mitigate Hallucinations in Large Vision-Language Models
机构 * New Laboratory of Pattern Recognition (NLPR), Institute of Automation, Chinese Academy of Sciences (CASIA)（模式识别新技术实验室，自动化研究所，中国科学院） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） ； Kuaishou Technology（快手科技） ； Nanjing University（南京大学）
Comments Accepted to Findings of ACL 2025
BiasGuard: A Reasoning-enhanced Bias Detection Tool For Large Language Models
机构 * Zhejiang University（浙江大学） ； Zhejiang Key Laboratory of Medical Imaging Artificial Intelligence（浙江省医学影像人工智能重点实验室）
Comments ACL 2025 findings
LLaSE-G1: Incentivizing Generalization Capability for LLaMA-based Speech Enhancement
机构 * Northwestern Polytechnical University（西北工业大学） ； The Hong Kong University of Science and Technology（香港科技大学） ； Huawei Technologies Co., Ltd.（华为技术有限公司）
Comments ACL2025 main, Codes available at https://github.com/Kevin-naticl/LLaSE-G1
Self-Training Elicits Concise Reasoning in Large Language Models
机构 * KAIST AI（韩国科学技术院人工智能研究所）
Comments 26 pages, 10 figures, 23 tables. Accepted to Findings of ACL 2025
Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence
机构 * Foundation Model Research Center, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所基础模型研究中心） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） ； National University of Singapore（新加坡国立大学） ； Southeast University（东南大学） ； Wuhan AI Research（武汉人工智能研究所）
Comments ACL2025
JuStRank: Benchmarking LLM Judges for System Ranking
机构 * IBM Research（IBM研究院）
Comments ACL 2025
DiffLM: Controllable Synthetic Data Generation via Diffusion Language Models
机构 * University of Chinese Academy of Sciences（中国科学院大学） ； Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所信息处理实验室） ； ByteDance Inc.（字节跳动公司）
Comments 21 pages, 9 figures, Accepted by ACL 2025, Findings
Length-Induced Embedding Collapse in PLM-based Models
机构 * Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学北京校区人工智能学院）
Comments Accepted by ACL 2025
Long-Context Language Modeling with Parallel Context Encoding
机构 * Princeton Language and Intelligence (PLI)（普林斯顿语言与智能中心）
Comments ACL 2024. Code, models, and data are available at https://github.com/princeton-nlp/CEPE
SciMON: Scientific Inspiration Machines Optimized for Novelty
Comments 21 pages. Code and resource are available at https://github.com/EagleW/CLBD Accepted by the 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024)
Multimedia Generative Script Learning for Task Planning
机构 * University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） ； University of Macau（澳门大学） ； Virginia Tech（弗吉尼亚理工大学）
Comments 21 pages, Accepted by Findings of the Association for Computational Linguistics: ACL 2023, Code and Resources at https://github.com/EagleW/Multimedia-Generative-Script-Learning
Stage-wise Fine-tuning for Graph-to-Text Generation
Comments 10 pages, Accepted by Proceedings of ACL-IJCNLP 2021 Student Research Workshop, Code and Resources at https://github.com/EagleW/Stage-wise-Fine-tuning
COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation
Comments 12 pages, Accepted by Proceedings of 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics System Demonstrations, for resources see http://blender.cs.illinois.edu/covid19/, for video see http://159.89.180.81/demo/covid/Covid-KG_DemoVideo.mp4, for slides see https://eaglew.github.io/files/Covid-KG_DemoVideo_with_ethics.pdf
Automatic Generation of Inference Making Questions for Reading Comprehension Assessments
机构 * Stanford University（斯坦福大学） ； ETS Research Institute（教育考试服务中心研究所）
Comments Accepted to the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2025), co-located with the ACL 2025
BehaviorBox: Automated Discovery of Fine-Grained Performance Differences Between Language Models
Comments Accepted to ACL 2025 Main Conference
Enhancing Open-Domain Task-Solving Capability of LLMs via Autonomous Tool Integration from GitHub
Comments Accepted by ACL 2025 Main Conference
