语言大模型 / LLM

ESI: Epistemic Uncertainty Quantification via Semantic-preserving Intervention for Large Language Models

Mingda Li, Xinyu Li, Weinan Zhang, Longxuan Ma

机构 * Harbin Institute of Technology（哈尔滨工业大学） ； Kunming University of Science and Technology（昆明理工大学） 纠错

专题命中 知识编辑与模型理解 ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI、cs.LG

Multi-Scale Probabilistic Generation Theory: A Unified Information-Theoretic Framework for Hierarchical Structure in Large Language Models

Yukin Zhang, Qi Dong

机构 * The Chinese University Of Hongkong（香港中文大学） ； Fudan University（复旦大学） ； University of Edinburgh（爱丁堡大学） 纠错

专题命中 知识编辑与模型理解 ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models

Anna Hedström, Salim I. Amoukou, Tom Bewley, Saumitra Mishra, Manuela Veloso

机构 * J.P. Morgan AI Research（摩根大通AI研究） 纠错

专题命中 知识编辑与模型理解 ：language model(title,abstract)；分类 cs.AI、cs.LG

Comments ICML 2025, 22 pages, 16 figures, 5 tables

Journal ref International Machine Learning Conference (ICML) 2025

A Matter of Representation: Towards Graph-Based Abstract Code Generation

Nyx Iskandar, Hisham Bedri, Andy Tsen

专题命中 知识编辑与模型理解 ：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

On the Limits of Language Generation: Trade-Offs Between Hallucination and Mode Collapse

Alkis Kalavasis, Anay Mehrotra, Grigoris Velegkas

机构 * Yale University（耶鲁大学） 纠错

专题命中 知识编辑与模型理解 ：language model(abstract)；post-training(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted for presentation at the 57th Symposium on Theory of Computing (STOC 2025); v3 fixes typos

Emergent Semantics Beyond Token Embeddings: Transformer LMs with Frozen Visual Unicode Representations

A. Bochkov

机构 * Moscow Institute of Physics and Technology (MIPT)（莫斯科物理技术学院） 纠错

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Published in Transactions on Machine Learning Research (10/2025). OpenReview: https://openreview.net/forum?id=Odh8IynO1o

Journal ref Transactions on Machine Learning Research (TMLR), 2025

Adaptive vector steering: A training-free, layer-wise intervention for hallucination mitigation in large audio and multimodal models

Tsung-En Lin, Kuan-Yi Lee, Hung-Yi Lee

机构 * National Taiwan University（国立台湾大学） ； ASUS Open Cloud Infrastructure Software Center（ASUS开放云基础设施软件中心） 纠错

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.LG

Comments Note: This preprint is a version of the paper submitted to ICASSP 2026. The author list here includes contributors who provided additional supervision and guidance. The official ICASSP submission may differ slightly in author composition

Estimating the Hallucination Rate of Generative AI

Andrew Jesson, Nicolas Beltran-Velez, Quentin Chu, Sweta Karlekar, Jannik Kossen, Yarin Gal, John P. Cunningham, David Blei

机构 * Department of Statistics, Columbia University（统计系，哥伦比亚大学） ； Department of Computer Science, Columbia University（计算机科学系，哥伦比亚大学） ； OATML, Department of Computer Science, University of Oxford（OATML，计算机科学系，牛津大学） 纠错

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.LG

PhysMaster: Mastering Physical Representation for Video Generation via Reinforcement Learning

Sihui Ji, Xi Chen, Xin Tao, Pengfei Wan, Hengshuang Zhao

机构 * The University of Hong Kong（香港大学） ； Kling Team, Kuaishou Technology（快手科技 Kling 团队） 纠错

专题命中 知识编辑与模型理解 ：preference optimization(abstract)

Comments Project Page: https://sihuiji.github.io/PhysMaster-Page/

QUIDS: Query Intent Description for Exploratory Search via Dual Space Modeling

Yumeng Wang, Xiuying Chen, Suzan Verberne

专题命中 知识编辑与模型理解 ：LLM(abstract)