EMNLP - arXivDaily 期刊&会议

2510.19493 2025-10-23 cs.CL

What is the Best Sequence Length for BABYLM?

Suchir Salhan, Richard Diehl Martinez, Zébulon Goriely, Paula Buttery

机构 * Department of Computer Science & Technology, University of Cambridge, U.K.（计算机科学与技术系，剑桥大学，英国）； ALTA Institute, University of Cambridge, U.K.（ALTA研究所，剑桥大学，英国）

Comments Paper Accepted at the 2025 BabyLM Workshop @ EMNLP (Suzhou, China)

2510.19419 2025-10-23 cs.CL

BLiSS 1.0: Evaluating Bilingual Learner Competence in Second Language Small Language Models

Yuan Gao, Suchir Salhan, Andrew Caines, Paula Buttery, Weiwei Sun

机构 * ALTA Institute（ALTA研究院）； Department of Computer Science & Technology, University of Cambridge（计算机科学与技术系，剑桥大学）

Comments Accepted Paper at the BabyLM Workshop 2025 @ EMNLP (Presentation in Suzhou, China)

2510.19005 2025-10-23 cs.CL

Dynamic Evaluation for Oversensitivity in LLMs

Sophia Xiao Pu, Sitao Cheng, Xin Eric Wang, William Yang Wang

机构 * University of California, Santa Barbara（加州大学圣芭芭拉分校）

Comments EMNLP-Findings 2025

2510.18279 2025-10-23 cs.CL cs.AI

Text or Pixels? It Takes Half: On the Token Efficiency of Visual Text Inputs in Multimodal LLMs

Yanhong Li, Zixuan Lan, Jiawei Zhou

机构 * Allen Institute for AI（艾伦人工智能研究所）； University of Chicago（芝加哥大学）； Stony Brook University（石溪大学）

Comments Accepted to EMNLP 2025 Findings ("Text or Pixels? Evaluating Efficiency and Understanding of LLMs with Visual Text Inputs")

2506.13579 2025-10-23 cs.LG cs.AI cs.CL cs.CV

Flexible-length Text Infilling for Discrete Diffusion Models

Andrew Zhang, Anushka Sivakumar, Chiawei Tang, Chris Thomas

机构 * Department of Computer Science（计算机科学系）

Comments Major edit of methodology section. Matches EMNLP camera-ready version

2505.15695 2025-10-23 cs.CL

Can Large Language Models be Effective Online Opinion Miners?

Ryang Heo, Yongsik Seo, Junseong Lee, Dongha Lee

机构 * Department of Artificial Intelligence, Yonsei University - DLI Lab（人工智能系，延世大学 - DLI实验室）

Comments Accepted to EMNLP 2025 Main

2503.12730 2025-10-23 cs.LG cs.AI cs.DB

TinySQL: A Progressive Text-to-SQL Dataset for Mechanistic Interpretability Research

Abir Harrasse, Philip Quirke, Clement Neo, Dhruv Nathawani, Luke Marks, Amir Abdullah

机构 * Martian ； Apart Research ； NVIDIA ； Thoughtworks ； Mohammed VI Polytechnic University（摩洛哥穆莱·易斯政院）

Comments Accepted to EMNLP 2025, 9 pages, 19 figures, 7 tables, 18 trained models Project Website: https://abirharrasse.github.io/tinysql/ Code Repository: https://github.com/withmartian/TinySQL Datasets and Models Viewer: https://huggingface.co/spaces/abir-hr196/tinysql-demo

2503.10620 2025-10-23 cs.CL

From TOWER to SPIRE: Adding the Speech Modality to a Translation-Specialist LLM

Kshitij Ambilduke, Ben Peters, Sonal Sannigrahi, Anil Keshwani, Tsz Kin Lam, Bruno Martins, André F. T. Martins, Marcely Zanon Boito

机构 * ENS Paris-Saclay（巴黎-萨克雷大学）； INESC-ID ； Instituto de Telecomunicações（电信研究所）； Instituto Superior Técnico, Universidade de Lisboa（里斯本大学技术学院）； Sapienza University of Rome（罗马萨皮恩扎大学）； University of Edinburgh（爱丁堡大学）； TransPerfect ； NAVER LABS Europe（NAVER欧洲实验室）

Comments EMNLP 2025 (Findings) camera ready

2503.07663 2025-10-23 cs.LG cs.AI

Merge then Realign: Simple and Effective Modality-Incremental Continual Learning for Multimodal LLMs

Dingkun Zhang, Shuhan Qi, Xinyu Xiao, Kehai Chen, Xuan Wang

机构 * Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））； Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies（广东省新型安全智能技术重点实验室）

Comments EMNLP 2025 Main Conference

2503.01872 2025-10-23 cs.LG cs.AI cs.CV

FairGen: Controlling Sensitive Attributes for Fair Generations in Diffusion Models via Adaptive Latent Guidance

Mintong Kang, Vinayshekhar Bannihatti Kumar, Shamik Roy, Abhishek Kumar, Sopan Khosla, Balakrishnan Murali Narayanaswamy, Rashmi Gangadharaiah

机构 * UIUC（伊利诺伊大学）； AWS AI Labs（亚马逊AI实验室）

Comments EMNLP 2025 Main Conference (Camera Ready)

2502.01349 2025-10-23 cs.CL

Bias Beware: The Impact of Cognitive Biases on LLM-Driven Product Recommendations

Giorgos Filandrianos, Angeliki Dimitriou, Maria Lymperaiou, Konstantinos Thomas, Giorgos Stamou

机构 * National Technical University of Athens（国家技术大学雅典）； Instituto de Telecomunicações（电信研究所）

Comments Accepted at EMNLP 2025

2501.04961 2025-10-23 cs.CL cs.AI cs.CE cs.LG

Demystifying Domain-adaptive Post-training for Financial LLMs

Zixuan Ke, Yifei Ming, Xuan-Phi Nguyen, Caiming Xiong, Shafiq Joty

机构 * Salesforce AI Research（Salesforce人工智能研究）

Comments EMNLP 2025 (Oral, ARR best paper nomination)

2411.16160 2025-10-23 cs.IR

Stop Playing the Guessing Game! Target-free User Simulation for Evaluating Conversational Recommender Systems

Sunghwan Kim, Kwangwook Seo, Tongyoung Kim, Jinyoung Yeo, Dongha Lee

Comments Accepted to EMNLP 2025 Findings