语言大模型 / LLM - arXivDaily 专题

2506.07233 2025-09-16 eess.AS cs.CL 79%

Reducing Object Hallucination in Large Audio-Language Models via Audio-Aware Decoding

Tzu-wen Hsu, Ke-Han Lu, Cheng-Han Chiang, Hung-yi Lee

机构 * Purdue University（普渡大学）； National Taiwan University（国立台湾大学）

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

2509.05186 2025-09-09 stat.ML cs.LG cs.NA math.NA 79%

Probabilistic operator learning: generative modeling and uncertainty quantification for foundation models of differential equations

Benjamin J. Zhang, Siting Liu, Stanley J. Osher, Markos A. Katsoulakis

机构 * Division of Applied Mathematics, Brown University（布朗大学应用数学系）； Department of Mathematics, University of California, Riverside（加州大学河滨分校数学系）； Department of Mathematics, University of California, Los Angeles（加州大学洛杉矶分校数学系）； Department of Mathematics and Statistics, University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校数学与统计学系）

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

Comments First two authors contributed equally

2509.05060 2025-09-08 cs.CL 79%

Entropy2Vec: Crosslingual Language Modeling Entropy as End-to-End Learnable Language Representations

Patrick Amadeus Irawan, Ryandito Diandaru, Belati Jagad Bintang Syuhada, Randy Zakya Suchrady, Alham Fikri Aji, Genta Indra Winata, Fajri Koto, Samuel Cahyawijaya

机构 * MBZUAI（马克斯·普朗克人工智能研究所）； Universitas Indonesia（印度尼西亚大学）； NTU（南洋理工大学）； Capital One（Capital One公司）； Cohere（Cohere公司）

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

2509.03816 2025-09-05 physics.ao-ph cs.LG 79%

Finetuning AI Foundation Models to Develop Subgrid-Scale Parameterizations: A Case Study on Atmospheric Gravity Waves

Aman Gupta, Aditi Sheshadri, Sujit Roy, Johannes Schmude, Vishal Gaur, Wei Ji Leong, Manil Maskey, Rahul Ramachandran

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

2509.02805 2025-09-04 cs.LG 79%

Challenges in Understanding Modality Conflict in Vision-Language Models

Trang Nguyen, Jackson Michaels, Madalina Fiterau, David Jensen

机构 * Manning College of Information \& Computer Sciences, University of Massachusetts Amherst, Amherst, U.S.

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

2509.00132 2025-09-03 cs.SD cs.AI cs.MM eess.AS 79%

CoComposer: LLM Multi-agent Collaborative Music Composition

Peiwen Xing, Aske Plaat, Niki van Stein

机构 * LIACS, Leiden University, Netherlands（莱顿大学莱顿信息与计算科学研究中心）

专题命中知识编辑与模型理解：LLM(title,abstract)；分类 cs.AI

2203.13722 2025-08-29 cs.CL 79%

Probing Pre-Trained Language Models for Cross-Cultural Differences in Values

Arnav Arora, Lucie-Aimée Kaffee, Isabelle Augenstein

机构 * University of Copenhagen（哥本哈根大学）

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

Comments Accepted to C3NLP, EACL 2023: https://aclanthology.org/2023.c3nlp-1.12/

2505.19498 2025-08-20 cs.CV cs.AI 79%

Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models

Nanxing Hu, Xiaoyue Duan, Jinchao Zhang, Guoliang Kang

机构 * Beihang University（北航大学）； Tencent WXG（腾讯 WXG）

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

2410.00332 2025-08-14 cs.AI q-bio.NC 79%

Vision Language Models Know Law of Conservation without Understanding More-or-Less

Dezhi Luo, Haiyun Lyu, Qingying Gao, Haoran Sun, Yijiang Li, Hokin Deng

机构 * University of Michigan（密歇根大学）； University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）； Johns Hopkins University（约翰霍普金斯大学）； University of California, San Diego（加州大学圣地亚哥分校）； Carnegie Mellon University（卡内基梅隆大学）

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

Comments Published at the ICLR 2025 Workshop on Bidirectional Human-AI Alignment (BiAlign)

2410.06795 2025-08-13 cs.CL cs.CV 79%

From Pixels to Tokens: Revisiting Object Hallucinations in Large Vision-Language Models

Yuying Shang, Xinyi Zeng, Yutao Zhu, Xiao Yang, Zhengwei Fang, Jingyuan Zhang, Jiawei Chen, Zinan Liu, Yu Tian

机构 * University of Chinese Academy of Sciences（中国科学院大学）； Dept. of Comp. Sci. and Tech., Institute for AI, Tsinghua University（计算机科学与技术系，人工智能研究院，清华大学）； Gaoling School of Artificial Intelligence, Renmin University of China（人工智能学院，中国人民大学）； Kuaishou Technology Inc.（快手科技有限公司）； Shanghai Key Laboratory of Multi. Info. Processing, East China Normal University（多信息处理重点实验室，华东师范大学）

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

2508.01678 2025-08-05 cs.CV cs.AI 79%

Cure or Poison? Embedding Instructions Visually Alters Hallucination in Vision-Language Models

Zhaochen Wang, Yiwei Wang, Yujun Cai

机构 * The University of Queensland（昆士兰大学）； University of California, Merced（加州大学默塞德分校）

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

Comments Work in progress

2507.05056 2025-07-23 cs.CV cs.AI 79%

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Xin Dong, Shichao Dong, Jin Wang, Jing Huang, Li Zhou, Zenghui Sun, Lihua Jing, Jingsong Lan, Xiaoyong Zhu, Bo Zheng

机构 * University of Chinese Academy of Sciences（中国科学院大学）； The University of Hong Kong（香港大学）； Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）

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

Comments Accepted by ICCV 2025

2507.14640 2025-07-22 cs.CL 79%

Linear Relational Decoding of Morphology in Language Models

Eric Xia, Jugal Kalita

机构 * Brown University（布朗大学）； University of Colorado Colorado Springs（科罗拉多州立大学）

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

Journal ref Proc. NAACL-HLT 2025 Student Research Workshop 4 (2025) 225-235

2406.17241 2025-07-17 cs.CL 79%

Understanding Language Model Circuits through Knowledge Editing

Huaizhi Ge, Frank Rudzicz, Zining Zhu

机构 * Columbia University（哥伦比亚大学）； Dalhousie University（达尔豪斯大学）； Stevens Institute of Technology（史蒂文斯理工学院）

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

Comments A previous version of this document contained a hidden prompt entered by Z Zhu without knowledge of -- or consent by -- his co-authors. This version does not contain the prompt

2402.08830 2025-07-14 cs.DS cs.CC cs.CL 79%

Sequence graphs realizations and ambiguity in language models

Sammy Khalife, Yann Ponty, Laurent Bulteau

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

2507.06458 2025-07-10 cs.LG q-bio.BM 79%

Automated Neuron Labelling Enables Generative Steering and Interpretability in Protein Language Models

Arjun Banerjee, David Martinez, Camille Dang, Ethan Tam

机构 * Department of Electrical Engineering and Computer Science, University of California, Berkeley, Berkeley, California（电气工程与计算机科学系，加州大学伯克利分校）

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

Comments 15 pages, 13 figures. Accepted to Proceedings of the Workshop on Generative AI for Biology at the 42nd International Conference on Machine Learning (Spotlight)

2505.05163 2025-07-08 cs.CV cs.LG 79%

Probabilistic Embeddings for Frozen Vision-Language Models: Uncertainty Quantification with Gaussian Process Latent Variable Models

Aishwarya Venkataramanan, Paul Bodesheim, Joachim Denzler

机构 * Computer Vision Group, Friedrich Schiller University Jena（计算机视觉组，费迪里奇·施勒尔大学耶纳）

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

Comments UAI 2025, 22 pages

2410.14405 2025-07-02 cs.CL 79%

Fact Recall, Heuristics or Pure Guesswork? Precise Interpretations of Language Models for Fact Completion

Denitsa Saynova, Lovisa Hagström, Moa Johansson, Richard Johansson, Marco Kuhlmann

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

Comments accepted to ACL Findings 2025

2506.21861 2025-06-30 cs.CL 79%

Derivational Probing: Unveiling the Layer-wise Derivation of Syntactic Structures in Neural Language Models

Taiga Someya, Ryo Yoshida, Hitomi Yanaka, Yohei Oseki

机构 * The University of Tokyo（东京大学）； RIKEN（日本研究机构）

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

2506.21468 2025-06-27 cs.CL 79%

TopK Language Models

Ryosuke Takahashi, Tatsuro Inaba, Kentaro Inui, Benjamin Heinzerling

机构 * Tohoku University（东北大学）； RIKEN（日本研究机构）； MBZUAI（多模态基础人工智能研究所）

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

2506.19513 2025-06-25 cs.CV cs.LG 79%

Visual hallucination detection in large vision-language models via evidential conflict

Tao Huang, Zhekun Liu, Rui Wang, Yang Zhang, Liping Jing

机构 * Beijing Key Lab of Traffic Data Mining（北京交通数据挖掘与具身智能重点实验室）； State Key Laboratory of Advanced Rail Autonomous Operation（先进轨道交通自主运行国家重点实验室）； School of Computer Science and Technology（计算机科学与技术学院）； Beijing Jiaotong University（北京交通大学）； School of Automation and Intelligence（自动化与智能学院）； School of Electronic and Information Engineering（电子与信息工程学院）

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

Journal ref International Journal of Approximate Reasoning, Volume 186, November 2025, Article 109507

2506.19498 2025-06-25 cs.RO cs.AI 79%

T-Rex: Task-Adaptive Spatial Representation Extraction for Robotic Manipulation with Vision-Language Models

Yiteng Chen, Wenbo Li, Shiyi Wang, Huiping Zhuang, Qingyao Wu

机构 * School of Software Engineering, South China University of Technology（软件工程学院，华南理工大学）； School of Future Technology, South China University of Technology（未来技术学院，华南理工大学）； Shien-Ming Wu School of Intelligent Engineering, South China University of Technology（智能工程学院，华南理工大学）

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

Comments submitted to NeurIPS 2025

2504.03036 2025-06-13 cs.CL 79%

IPA-CHILDES & G2P+: Feature-Rich Resources for Cross-Lingual Phonology and Phonemic Language Modeling

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）； ALTA Institute, University of Cambridge, U.K.（ALTA研究所，剑桥大学）

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

Comments Accepted to CoNLL 2025

2506.09593 2025-06-12 cs.LG 79%

Beyond Overconfidence: Foundation Models Redefine Calibration in Deep Neural Networks

Achim Hekler, Lukas Kuhn, Florian Buettner

机构 * Goethe University Frankfurt（弗赖堡歌德大学）； German Cancer Consortium (DKTK)（德国癌症联盟（DKTK））； German Cancer Research Center (DKFZ)（德国癌症研究中心（DKFZ））； Frankfurt Cancer Institute（法兰克福癌症研究所）

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

2506.06686 2025-06-10 cs.CL 79%

Learning Distribution-Wise Control in Representation Space for Language Models

Chunyuan Deng, Ruidi Chang, Hanjie Chen

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

Comments ICML 2025

2506.05136 2025-06-06 cs.CL 79%

Information Locality as an Inductive Bias for Neural Language Models

Taiga Someya, Anej Svete, Brian DuSell, Timothy J. O'Donnell, Mario Giulianelli, Ryan Cotterell

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

2502.12414 2025-06-06 cs.CL 79%

Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models

Hanin Atwany, Abdul Waheed, Rita Singh, Monojit Choudhury, Bhiksha Raj

机构 * Carnegie Mellon University（卡内基梅隆大学）； MBZUAI（穆扎布伊人工智能研究所）

专题命中知识编辑与模型理解：foundation model(title,abstract)；分类 cs.CL

Comments ACL2025 camera-ready

2506.03741 2025-06-05 cs.HC cs.CL 79%

PromptCanvas: Composable Prompting Workspaces Using Dynamic Widgets for Exploration and Iteration in Creative Writing

Rifat Mehreen Amin, Oliver Hans Kühle, Daniel Buschek, Andreas Butz

机构 * LMU Munich（慕尼黑莱布尼茨大学）； University of Bayreuth（拜罗伊特大学）

专题命中知识编辑与模型理解：prompting(title,abstract)；分类 cs.CL

2505.24731 2025-06-02 cs.CL 79%

Circuit Stability Characterizes Language Model Generalization

Alan Sun

机构 * Carnegie Mellon University（卡内基梅隆大学）

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

Comments 16 pages, 10 figures

2505.24649 2025-06-02 cs.CV cs.AI 79%

BIMA: Bijective Maximum Likelihood Learning Approach to Hallucination Prediction and Mitigation in Large Vision-Language Models

Huu-Thien Tran, Thanh-Dat Truong, Khoa Luu

机构 * CVIU Lab, University of Arkansas（CVIU实验室，阿肯色大学）

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

Comments CVPRW 2025, 8 pages, 4 figures

1. 知识编辑与模型理解 7596 篇

Reducing Object Hallucination in Large Audio-Language Models via Audio-Aware Decoding

Probabilistic operator learning: generative modeling and uncertainty quantification for foundation models of differential equations

Entropy2Vec: Crosslingual Language Modeling Entropy as End-to-End Learnable Language Representations

Finetuning AI Foundation Models to Develop Subgrid-Scale Parameterizations: A Case Study on Atmospheric Gravity Waves

Challenges in Understanding Modality Conflict in Vision-Language Models

CoComposer: LLM Multi-agent Collaborative Music Composition

Probing Pre-Trained Language Models for Cross-Cultural Differences in Values

Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models

Vision Language Models Know Law of Conservation without Understanding More-or-Less

From Pixels to Tokens: Revisiting Object Hallucinations in Large Vision-Language Models

Cure or Poison? Embedding Instructions Visually Alters Hallucination in Vision-Language Models

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Linear Relational Decoding of Morphology in Language Models

Understanding Language Model Circuits through Knowledge Editing

Sequence graphs realizations and ambiguity in language models

Automated Neuron Labelling Enables Generative Steering and Interpretability in Protein Language Models

Probabilistic Embeddings for Frozen Vision-Language Models: Uncertainty Quantification with Gaussian Process Latent Variable Models

Fact Recall, Heuristics or Pure Guesswork? Precise Interpretations of Language Models for Fact Completion

Derivational Probing: Unveiling the Layer-wise Derivation of Syntactic Structures in Neural Language Models

TopK Language Models

Visual hallucination detection in large vision-language models via evidential conflict

T-Rex: Task-Adaptive Spatial Representation Extraction for Robotic Manipulation with Vision-Language Models

IPA-CHILDES & G2P+: Feature-Rich Resources for Cross-Lingual Phonology and Phonemic Language Modeling

Beyond Overconfidence: Foundation Models Redefine Calibration in Deep Neural Networks

Learning Distribution-Wise Control in Representation Space for Language Models

Information Locality as an Inductive Bias for Neural Language Models

Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models

PromptCanvas: Composable Prompting Workspaces Using Dynamic Widgets for Exploration and Iteration in Creative Writing

Circuit Stability Characterizes Language Model Generalization

BIMA: Bijective Maximum Likelihood Learning Approach to Hallucination Prediction and Mitigation in Large Vision-Language Models