语言大模型 / LLM - arXivDaily 专题

2505.24244 2025-06-02 cs.CL cs.LG 62%

Mamba Knockout for Unraveling Factual Information Flow

Nir Endy, Idan Daniel Grosbard, Yuval Ran-Milo, Yonatan Slutzky, Itay Tshuva, Raja Giryes

机构 * Tel Aviv University（特拉维夫大学）

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

Comments Accepted to ACL 2025

2505.22803 2025-05-30 cs.LG cs.AI 62%

CLUE: Neural Networks Calibration via Learning Uncertainty-Error alignment

Pedro Mendes, Paolo Romano, David Garlan

机构 * Software and Societal Systems Department, Carnegie Mellon University（卡内基梅隆大学软件与社会系统部门）； INESC-ID and Instituto Superior Técnico, Universidade de Lisboa（里斯本大学INESC-ID和理工学院）

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

2410.15268 2025-05-30 cs.LG cs.CL 62%

GraphNarrator: Generating Textual Explanations for Graph Neural Networks

Bo Pan, Zhen Xiong, Guanchen Wu, Zheng Zhang, Yifei Zhang, Liang Zhao

机构 * Emory University（埃默里大学）

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

Comments ACL 2025 (Main)

2502.20578 2025-05-29 cs.CV cs.AI cs.LG 62%

Interpreting CLIP with Hierarchical Sparse Autoencoders

Vladimir Zaigrajew, Hubert Baniecki, Przemyslaw Biecek

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

Journal ref Proceedings of the 42st International Conference on Machine Learning (ICML 2025)

2502.09245 2025-05-29 cs.LG cs.CL 62%

You Do Not Fully Utilize Transformer's Representation Capacity

Gleb Gerasimov, Yaroslav Aksenov, Nikita Balagansky, Viacheslav Sinii, Daniil Gavrilov

机构 * T-Tech ； Moscow Institute of Physics and Technology（莫斯科物理技术学院）； HSE University（俄罗斯高等经济大学）

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

2505.17936 2025-05-26 cs.LG cs.CL 62%

Understanding Gated Neurons in Transformers from Their Input-Output Functionality

Sebastian Gerstner, Hinrich Schütze

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

Comments 31 pages, 22 figures

2502.05300 2025-05-26 cs.LG cond-mat.dis-nn cs.AI stat.ML 62%

Parameter Symmetry Potentially Unifies Deep Learning Theory

Liu Ziyin, Yizhou Xu, Tomaso Poggio, Isaac Chuang

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

Comments preprint

2502.20167 2025-05-23 cs.LG cs.CL 62%

Similarity-Distance-Magnitude Universal Verification

Allen Schmaltz

机构 * Reexpress AI

专题命中知识编辑与模型理解：LLM(abstract)；分类 cs.CL、cs.LG

Comments 36 pages (1 Figure, 8 Tables, 4 Algorithms, 5 Listings)

2505.13210 2025-05-20 cs.CL cs.AI 62%

Picturized and Recited with Dialects: A Multimodal Chinese Representation Framework for Sentiment Analysis of Classical Chinese Poetry

Xiaocong Du, Haoyu Pei, Haipeng Zhang

机构 * ShanghaiTech University（上海科技大学）

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

2505.11836 2025-05-20 cs.LG cs.AI 62%

SplInterp: Improving our Understanding and Training of Sparse Autoencoders

Jeremy Budd, Javier Ideami, Benjamin Macdowall Rynne, Keith Duggar, Randall Balestriero

机构 * School of Mathematics University of Birmingham（数学学院英国伯明翰大学）； Ideami Studios（Ideami工作室）； Department of Mathematics and Statistics University of Limerick（数学与统计学学院英国利默里克大学）； XRAI Inc.（XRAI公司）； Department of Computer Science Brown University（计算机科学系布朗大学）

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

Comments 44 pages, 38 figures, under review

2402.03607 2025-05-19 cs.AI cs.CL cs.CV cs.CY cs.HC 62%

Enhancing Cross-Modal Contextual Congruence for Crowdfunding Success using Knowledge-infused Learning

Trilok Padhi, Ugur Kursuncu, Yaman Kumar, Valerie L. Shalin, Lane Peterson Fronczek

机构 * Georgia State University（佐治亚州立大学）； Adobe MDSR ； Wright State University（怀特州立大学）； California Polytechnic State University（加州州立大学帕克校区）

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

Comments Accepted at IEEE International Conference on Big Data 2024 (IEEE BigData 2024)

Journal ref IEEE International Conference on Big Data 2024 (IEEE BigData 2024)

2505.06032 2025-05-12 cs.LG cs.CL 62%

Short-circuiting Shortcuts: Mechanistic Investigation of Shortcuts in Text Classification

Leon Eshuijs, Shihan Wang, Antske Fokkens

机构 * Vrije Universiteit Amsterdam（阿姆斯特丹自由大学）； Utrecht University（乌特雷赫大学）

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

2412.16086 2025-04-30 cs.IR cs.AI cs.CL cs.CV eess.IV 62%

Towards Interpretable Radiology Report Generation via Concept Bottlenecks using a Multi-Agentic RAG

Hasan Md Tusfiqur Alam, Devansh Srivastav, Md Abdul Kadir, Daniel Sonntag

机构 * German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心（DFKI））； University of Oldenburg（奥尔登堡大学）

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

Comments Accepted in the 47th European Conference for Information Retrieval (ECIR) 2025

Journal ref Lecture Notes in Computer Science (LNCS) 2025, Volume 15574

2403.14562 2025-04-30 cs.CL cs.AI cs.HC cs.MA 62%

Agentic AI: The Era of Semantic Decoding

Maxime Peyrard, Martin Josifoski, Robert West

机构 * Univ. Grenoble Alpes, CNRS, Grenoble INP, LIG EPFL（格勒诺布尔阿尔卑斯大学、国家科学研究中心、格勒诺布尔INP、LIG EPFL）

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

Comments 25 pages, 3 figures

2504.18872 2025-04-29 cs.CL cs.LG 62%

Latent Adversarial Training Improves the Representation of Refusal

Alexandra Abbas, Nora Petrova, Helios Ael Lyons, Natalia Perez-Campanero

机构 * Apart Research（Apart研究）

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

2504.17261 2025-04-25 cs.LG cs.AI 62%

Symbolic Representation for Any-to-Any Generative Tasks

Jiaqi Chen, Xiaoye Zhu, Yue Wang, Tianyang Liu, Xinhui Chen, Ying Chen, Chak Tou Leong, Yifei Ke, Joseph Liu, Yiwen Yuan, Julian McAuley, Li-jia Li

机构 * Stanford University（斯坦福大学）； Fellou AI（Fellou人工智能）； Fudan University（复旦大学）； South China University of Technology（华南理工大学）； Cornell University（康奈尔大学）； University of California San Diego（加州大学圣地亚哥分校）； Wuhan University（武汉大学）； University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）； Hong Kong Polytechnic University（香港理工大学）； University of Southern California（南加州大学）； Carnegie Mellon University（卡内基梅隆大学）； LiveX AI（LiveX人工智能）

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

2408.12022 2025-04-21 cs.CL cs.AI 62%

Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind

Lance Ying, Tan Zhi-Xuan, Lionel Wong, Vikash Mansinghka, Joshua B. Tenenbaum

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

Comments 23 pages; Published at the Transactions of the Association for Computational Linguistics (TACL); Presented at NAACL 2025

2411.04011 2025-04-18 eess.SY cs.AI cs.LG cs.SY 62%

Predicting and Publishing Accurate Imbalance Prices Using Monte Carlo Tree Search

Fabio Pavirani, Jonas Van Gompel, Seyed Soroush Karimi Madahi, Bert Claessens, Chris Develder

专题命中知识编辑与模型理解：prompting(abstract)；分类 cs.AI、cs.LG

2504.09704 2025-04-15 cs.LG cs.AI 62%

Transformer-Based Representation Learning for Robust Gene Expression Modeling and Cancer Prognosis

Shuai Jiang, Saeed Hassanpour

专题命中知识编辑与模型理解：pretraining(abstract)；分类 cs.AI、cs.LG

2402.06038 2025-04-11 cs.LG cs.AI cs.CV 62%

Understanding Contrastive Representation Learning from Positive Unlabeled (PU) Data

Anish Acharya, Li Jing, Bhargav Bhushanam, Dhruv Choudhary, Michael Rabbat, Sujay Sanghavi, Inderjit S Dhillon

专题命中知识编辑与模型理解：SFT(abstract)；分类 cs.AI、cs.LG

2504.03857 2025-04-08 cs.CV cs.AI cs.LG 62%

Can ChatGPT Learn My Life From a Week of First-Person Video?

Keegan Harris

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

2410.04315 2025-04-03 cs.CL cs.LG 62%

Calibrating Expressions of Certainty

Peiqi Wang, Barbara D. Lam, Yingcheng Liu, Ameneh Asgari-Targhi, Rameswar Panda, William M. Wells, Tina Kapur, Polina Golland

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

Comments International Conference on Learning Representations (ICLR), 2025

2503.24228 2025-04-01 cs.AI cs.CL cs.MA 62%

PAARS: Persona Aligned Agentic Retail Shoppers

Saab Mansour, Leonardo Perelli, Lorenzo Mainetti, George Davidson, Stefano D'Amato

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

2503.23819 2025-04-01 cs.LG cs.AI cs.CV 62%

Conformal uncertainty quantification to evaluate predictive fairness of foundation AI model for skin lesion classes across patient demographics

Swarnava Bhattacharyya, Umapada Pal, Tapabrata Chakraborti

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

2503.23615 2025-04-01 cs.AI cs.LG 62%

An Organizationally-Oriented Approach to Enhancing Explainability and Control in Multi-Agent Reinforcement Learning

Julien Soulé, Jean-Paul Jamont, Michel Occello, Louis-Marie Traonouez, Paul Théron

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

2503.19900 2025-03-26 cs.CV cs.AI cs.CL 62%

CAFe: Unifying Representation and Generation with Contrastive-Autoregressive Finetuning

Hao Yu, Zhuokai Zhao, Shen Yan, Lukasz Korycki, Jianyu Wang, Baosheng He, Jiayi Liu, Lizhu Zhang, Xiangjun Fan, Hanchao Yu

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

2405.19186 2025-03-26 cs.CV cs.CL cs.LG 62%

MetaToken: Detecting Hallucination in Image Descriptions by Meta Classification

Laura Fieback, Jakob Spiegelberg, Hanno Gottschalk

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

2503.18551 2025-03-25 cs.LG cs.AI 62%

Discriminative protein sequence modelling with Latent Space Diffusion

Eoin Quinn, Ghassene Jebali, Maxime Seince, Oliver Bent

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

2407.20899 2025-03-19 cs.AI cs.CL 62%

Faithful and Plausible Natural Language Explanations for Image Classification: A Pipeline Approach

Adam Wojciechowski, Mateusz Lango, Ondrej Dusek

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

Comments Findings of EMNLP 2024

2412.07658 2025-03-18 cs.CV cs.AI cs.LG 62%

TraSCE: Trajectory Steering for Concept Erasure

Anubhav Jain, Yuya Kobayashi, Takashi Shibuya, Yuhta Takida, Nasir Memon, Julian Togelius, Yuki Mitsufuji

专题命中知识编辑与模型理解：prompting(abstract)；分类 cs.AI、cs.LG

1. 知识编辑与模型理解 7565 篇

Mamba Knockout for Unraveling Factual Information Flow

CLUE: Neural Networks Calibration via Learning Uncertainty-Error alignment

GraphNarrator: Generating Textual Explanations for Graph Neural Networks

Interpreting CLIP with Hierarchical Sparse Autoencoders

You Do Not Fully Utilize Transformer's Representation Capacity

Understanding Gated Neurons in Transformers from Their Input-Output Functionality

Parameter Symmetry Potentially Unifies Deep Learning Theory

Similarity-Distance-Magnitude Universal Verification

Picturized and Recited with Dialects: A Multimodal Chinese Representation Framework for Sentiment Analysis of Classical Chinese Poetry

SplInterp: Improving our Understanding and Training of Sparse Autoencoders

Enhancing Cross-Modal Contextual Congruence for Crowdfunding Success using Knowledge-infused Learning

Short-circuiting Shortcuts: Mechanistic Investigation of Shortcuts in Text Classification

Towards Interpretable Radiology Report Generation via Concept Bottlenecks using a Multi-Agentic RAG

Agentic AI: The Era of Semantic Decoding

Latent Adversarial Training Improves the Representation of Refusal

Symbolic Representation for Any-to-Any Generative Tasks

Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind

Predicting and Publishing Accurate Imbalance Prices Using Monte Carlo Tree Search

Transformer-Based Representation Learning for Robust Gene Expression Modeling and Cancer Prognosis

Understanding Contrastive Representation Learning from Positive Unlabeled (PU) Data

Can ChatGPT Learn My Life From a Week of First-Person Video?

Calibrating Expressions of Certainty

PAARS: Persona Aligned Agentic Retail Shoppers

Conformal uncertainty quantification to evaluate predictive fairness of foundation AI model for skin lesion classes across patient demographics

An Organizationally-Oriented Approach to Enhancing Explainability and Control in Multi-Agent Reinforcement Learning

CAFe: Unifying Representation and Generation with Contrastive-Autoregressive Finetuning

MetaToken: Detecting Hallucination in Image Descriptions by Meta Classification

Discriminative protein sequence modelling with Latent Space Diffusion

Faithful and Plausible Natural Language Explanations for Image Classification: A Pipeline Approach

TraSCE: Trajectory Steering for Concept Erasure