大模型推理能力 - arXivDaily 专题

2507.08637 2025-07-14 cs.LG cs.AI cs.CL 67%

Scaling Attention to Very Long Sequences in Linear Time with Wavelet-Enhanced Random Spectral Attention (WERSA)

Vincenzo Dentamaro

机构 * Department of Computer Science University of Bari Aldo Moro（计算机科学系巴里大学Aldo Moro）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 10 pages, 1 figure

2501.09879 2025-07-10 cs.SE 67%

Testing Refactoring Engine via Historical Bug Report driven LLM

Haibo Wang, Zhuolin Xu, Shin Hwei Tan

专题命中其他推理：chain-of-thought(abstract)；CoT(abstract)

Comments Accepted at the 2nd ACM international conference on AI Foundation Models and Software Engineering (FORGE 2025)

2507.04069 2025-07-08 cs.CL cs.AI cs.LG 67%

Beyond Independent Passages: Adaptive Passage Combination Retrieval for Retrieval Augmented Open-Domain Question Answering

Ting-Wen Ko, Jyun-Yu Jiang, Pu-Jen Cheng

机构 * National Taiwan University（国立台湾大学）； Amazon Search（亚马逊搜索）； University College London（伦敦大学学院）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2507.02822 2025-07-04 cs.CL cs.AI cs.LG 67%

SynapseRoute: An Auto-Route Switching Framework on Dual-State Large Language Model

Wencheng Zhang, Shiqin Qiao, Lingjie Luo, Yinfeng Li, Chuanyang Zheng, Qian Xu, Meng Li, Yong Gui, Yijun He, Jianing Qiu, Jindong Hong, Jiankai Sun

机构 * Bytedance（字节跳动）； Xidian University（西安电子科技大学）； The Chinese University of Hong Kong（香港中文大学）； Peking University（北京大学）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2503.04722 2025-07-01 cs.CL cs.AI cs.LG 67%

Enough Coin Flips Can Make LLMs Act Bayesian

Ritwik Gupta, Rodolfo Corona, Jiaxin Ge, Eric Wang, Dan Klein, Trevor Darrell, David M. Chan

机构 * University of California, Berkeley（加州大学伯克利分校）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments ACL 2025 Main

2502.04397 2025-07-01 cs.CL cs.AI cs.LG 67%

Multimodal Medical Code Tokenizer

Xiaorui Su, Shvat Messica, Yepeng Huang, Ruth Johnson, Lukas Fesser, Shanghua Gao, Faryad Sahneh, Marinka Zitnik

机构 * Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA（生物医学信息学系，哈佛医学院，波士顿，马萨诸塞州，美国）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments ICML'25

2411.08870 2025-07-01 cs.CL cs.AI cs.LG 67%

The Limited Impact of Medical Adaptation of Large Language and Vision-Language Models

Daniel P. Jeong, Pranav Mani, Saurabh Garg, Zachary C. Lipton, Michael Oberst

机构 * Carnegie Mellon University（卡内基梅隆大学）； Abridge ； Mistral AI ； Johns Hopkins University（约翰霍普金斯大学）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Extended version of EMNLP 2024 paper arXiv:2411.04118. Includes additional results on clinical note QA tasks and supervised fine-tuning evaluations

2506.14794 2025-06-19 cs.LG cs.AI cs.CL 67%

Assembly of Experts: Linear-time construction of the Chimera LLM variants with emergent and adaptable behaviors

Henrik Klagges, Robert Dahlke, Fabian Klemm, Benjamin Merkel, Daniel Klingmann, David A. Reiss, Dan Zecha

机构 * TNG Technology Consulting GmbH（TNG技术咨询公司）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2506.14280 2025-06-18 cs.LG cs.AI cs.CL stat.ML 67%

Improving LoRA with Variational Learning

Bai Cong, Nico Daheim, Yuesong Shen, Rio Yokota, Mohammad Emtiyaz Khan, Thomas Möllenhoff

机构 * Institute of Science Tokyo（东京科学研究院）； RIKEN Center for AI Project（RIKEN人工智能项目研究中心）； Ubiquitous Knowledge Processing Lab (UKP Lab), Department of Computer Science, Technical University of Darmstadt（通用知识处理实验室（UKP实验室），计算机科学系，达姆施塔特技术大学）； National Research Center for Applied Cybersecurity ATHENE, Germany（应用网络安全国家研究中心ATHENE，德国）； Huawei Hong Kong Research Center（华为香港研究中心）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 16 pages, 4 figures

2506.07833 2025-06-16 cs.LG cs.AI cs.CL 67%

Improving Large Language Models with Concept-Aware Fine-Tuning

Michael K. Chen, Xikun Zhang, Jiaxing Huang, Dacheng Tao

机构 * Nanyang Technological University（南洋理工大学）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2412.16187 2025-06-06 cs.LG cs.AI cs.CL cs.DS cs.PF 67%

HashEvict: A Pre-Attention KV Cache Eviction Strategy using Locality-Sensitive Hashing

Minghui Liu, Tahseen Rabbani, Tony O'Halloran, Ananth Sankaralingam, Mary-Anne Hartley, Furong Huang, Cornelia Fermüller, Yiannis Aloimonos

机构 * University of Maryland（马里兰大学）； Yale University（耶鲁大学）； University of Galway（Galway大学）； University of Maryland Capital One（马里兰大学Capital One）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 10 pages, 6 figures, 2 tables

2410.12872 2025-06-06 cs.CL cs.AI cs.CY cs.LG 67%

Not All Options Are Created Equal: Textual Option Weighting for Token-Efficient LLM-Based Knowledge Tracing

JongWoo Kim, SeongYeub Chu, Bryan Wong, Mun Yi

机构 * KAIST（韩国科学技术院）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 11 pages

2505.20322 2025-06-04 cs.CL cs.AI cs.CV cs.IR cs.LG 67%

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms

Mengru Wang, Ziwen Xu, Shengyu Mao, Shumin Deng, Zhaopeng Tu, Huajun Chen, Ningyu Zhang

机构 * Zhejiang University（浙江大学）； Tencent（腾讯）； National University of Singapore（新加坡国立大学）； NUS-NCS Joint Lab（新加坡国立大学NCS联合实验室）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments ACL 2025

2505.18458 2025-06-03 cs.DB cs.AI cs.CL cs.IR cs.LG 67%

A Survey of LLM $\times$ DATA

Xuanhe Zhou, Junxuan He, Wei Zhou, Haodong Chen, Zirui Tang, Haoyu Zhao, Xin Tong, Guoliang Li, Youmin Chen, Jun Zhou, Zhaojun Sun, Binyuan Hui, Shuo Wang, Conghui He, Zhiyuan Liu, Jingren Zhou, Fan Wu

机构 * Shanghai Jiao Tong University（上海交通大学）； Tsinghua University（清华大学）； Alibaba Group（阿里巴巴集团）； Shanghai AI Laboratory（上海人工智能实验室）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Please refer to the paper list at: https://github.com/weAIDB/awesome-data-llm

2502.01179 2025-05-30 cs.CL cs.AI cs.LG 67%

Joint Localization and Activation Editing for Low-Resource Fine-Tuning

Wen Lai, Alexander Fraser, Ivan Titov

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted by ICML 2025 (camera-ready version). The code is released at https://github.com/wenlai-lavine/jola

2502.13458 2025-05-29 cs.CL cs.AI cs.CR cs.LG 67%

ThinkGuard: Deliberative Slow Thinking Leads to Cautious Guardrails

Xiaofei Wen, Wenxuan Zhou, Wenjie Jacky Mo, Muhao Chen

机构 * University of California, Davis（加州大学戴维斯分校）； University of Southern California（南加州大学）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments ACL 2025

2505.18412 2025-05-27 cs.CV cs.HC 67%

Rehabilitation Exercise Quality Assessment and Feedback Generation Using Large Language Models with Prompt Engineering

Jessica Tang, Ali Abedi, Tracey J. F. Colella, Shehroz S. Khan

机构 * KITE Research Institute（KITE研究 institute）； Toronto Rehabilitation Institute（多伦多康复研究所）； University Health Network（大学健康网络）； Faculty of Applied Science and Engineering（应用科学与工程学院）； University of Toronto（多伦多大学）； College of Engineering and Technology（工程与技术学院）； American University of the Middle East（中东美国大学）

专题命中其他推理：reasoning(abstract)；chain-of-thought(abstract)

Comments 16 pages, 3 figures, 5 tables

2410.13852 2025-05-22 cs.CL cs.AI cs.CV cs.LG 67%

Retrospective Learning from Interactions

Zizhao Chen, Mustafa Omer Gul, Yiwei Chen, Gloria Geng, Anne Wu, Yoav Artzi

机构 * Department of Computer Science and Cornell Tech, Cornell University（计算机科学系和康奈尔科技学院，康奈尔大学）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2504.19627 2025-05-20 cs.CL cs.AI cs.CV cs.LG 67%

VCM: Vision Concept Modeling Based on Implicit Contrastive Learning with Vision-Language Instruction Fine-Tuning

Run Luo, Renke Shan, Longze Chen, Ziqiang Liu, Lu Wang, Min Yang, Xiaobo Xia

机构 * Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究所）； University of Chinese Academy of Sciences（中国科学院大学）； National University of Singapore（新加坡国立大学）； MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition, University of Science and Technology of China（中国科学技术大学脑启发智能感知与认知教育部重点实验室）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments VCM

2411.03883 2025-04-24 cs.CL cs.AI cs.LG 67%

MEG: Medical Knowledge-Augmented Large Language Models for Question Answering

Laura Cabello, Carmen Martin-Turrero, Uchenna Akujuobi, Anders Søgaard, Carlos Bobed

机构 * University of Copenhagen（哥本哈根大学）； Sony AI（索尼人工智能）； University of Zaragoza（阿拉维萨大学）

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2405.01744 2025-04-18 cs.LG cs.AI cs.CL stat.ME 67%

ALCM: Autonomous LLM-Augmented Causal Discovery Framework

Elahe Khatibi, Mahyar Abbasian, Zhongqi Yang, Iman Azimi, Amir M. Rahmani

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2407.12879 2025-04-17 cs.CL cs.AI cs.LG 67%

Large Visual-Language Models Are Also Good Classifiers: A Study of In-Context Multimodal Fake News Detection

Ye Jiang, Yimin Wang

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Withdraw for new experiments

2004.13821 2025-04-15 cs.CL cs.AI cs.LG 67%

Fine-tuning Multi-hop Question Answering with Hierarchical Graph Network

Guanming Xiong

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Incomplete Work

2501.15140 2025-04-01 cs.CV cs.AI cs.CL cs.LG 67%

Analyzing and Boosting the Power of Fine-Grained Visual Recognition for Multi-modal Large Language Models

Hulingxiao He, Geng Li, Zijun Geng, Jinglin Xu, Yuxin Peng

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Published as a conference paper at ICLR 2025. The model is available at https://huggingface.co/StevenHH2000/Finedefics

2410.10870 2025-04-01 cs.CL cs.AI cs.LG 67%

PortLLM: Personalizing Evolving Large Language Models with Training-Free and Portable Model Patches

Rana Muhammad Shahroz Khan, Pingzhi Li, Sukwon Yun, Zhenyu Wang, Shahriar Nirjon, Chau-Wai Wong, Tianlong Chen

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

2411.07279 2025-03-26 cs.AI cs.CL cs.LG 67%

The Surprising Effectiveness of Test-Time Training for Few-Shot Learning

Ekin Akyürek, Mehul Damani, Adam Zweiger, Linlu Qiu, Han Guo, Jyothish Pari, Yoon Kim, Jacob Andreas

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Preprint

2412.00876 2025-03-24 cs.CV cs.AI cs.CL cs.LG 67%

Dynamic-LLaVA: Efficient Multimodal Large Language Models via Dynamic Vision-language Context Sparsification

Wenxuan Huang, Zijie Zhai, Yunhang Shen, Shaosheng Cao, Fei Zhao, Xiangfeng Xu, Zheyu Ye, Yao Hu, Shaohui Lin

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted to ICLR 2025. Code is available at https://github.com/Osilly/dynamic_llava

2406.11402 2025-03-13 cs.CL cs.AI cs.LG 67%

Are Small Language Models Ready to Compete with Large Language Models for Practical Applications?

Neelabh Sinha, Vinija Jain, Aman Chadha

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted at The Fifth Workshop on Trustworthy Natural Language Processing (TrustNLP 2025) in Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), 2025. 8 pages + references + Appendix

2406.08155 2025-02-26 cs.LG cs.AI cs.CL 67%

QuantMoE-Bench: Examining Post-Training Quantization for Mixture-of-Experts

Pingzhi Li, Xiaolong Jin, Zhen Tan, Yu Cheng, Tianlong Chen

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Our code for reproducing all our experiments is provided at https://github.com/UNITES-Lab/moe-quantization

2502.15696 2025-02-25 cs.CL cs.AI cs.IR cs.LG 67%

Integrating Domain Knowledge into Large Language Models for Enhanced Fashion Recommendations

Zhan Shi, Shanglin Yang

专题命中其他推理：reasoning(abstract)；分类 cs.CL、cs.AI、cs.LG

1. 其他推理 5804 篇

Scaling Attention to Very Long Sequences in Linear Time with Wavelet-Enhanced Random Spectral Attention (WERSA)

Testing Refactoring Engine via Historical Bug Report driven LLM

Beyond Independent Passages: Adaptive Passage Combination Retrieval for Retrieval Augmented Open-Domain Question Answering

SynapseRoute: An Auto-Route Switching Framework on Dual-State Large Language Model

Enough Coin Flips Can Make LLMs Act Bayesian

Multimodal Medical Code Tokenizer

The Limited Impact of Medical Adaptation of Large Language and Vision-Language Models

Assembly of Experts: Linear-time construction of the Chimera LLM variants with emergent and adaptable behaviors

Improving LoRA with Variational Learning

Improving Large Language Models with Concept-Aware Fine-Tuning

HashEvict: A Pre-Attention KV Cache Eviction Strategy using Locality-Sensitive Hashing

Not All Options Are Created Equal: Textual Option Weighting for Token-Efficient LLM-Based Knowledge Tracing

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms

A Survey of LLM $\times$ DATA

Joint Localization and Activation Editing for Low-Resource Fine-Tuning

ThinkGuard: Deliberative Slow Thinking Leads to Cautious Guardrails

Rehabilitation Exercise Quality Assessment and Feedback Generation Using Large Language Models with Prompt Engineering

Retrospective Learning from Interactions

VCM: Vision Concept Modeling Based on Implicit Contrastive Learning with Vision-Language Instruction Fine-Tuning

MEG: Medical Knowledge-Augmented Large Language Models for Question Answering

ALCM: Autonomous LLM-Augmented Causal Discovery Framework

Large Visual-Language Models Are Also Good Classifiers: A Study of In-Context Multimodal Fake News Detection

Fine-tuning Multi-hop Question Answering with Hierarchical Graph Network

Analyzing and Boosting the Power of Fine-Grained Visual Recognition for Multi-modal Large Language Models

PortLLM: Personalizing Evolving Large Language Models with Training-Free and Portable Model Patches

The Surprising Effectiveness of Test-Time Training for Few-Shot Learning

Dynamic-LLaVA: Efficient Multimodal Large Language Models via Dynamic Vision-language Context Sparsification

Are Small Language Models Ready to Compete with Large Language Models for Practical Applications?

QuantMoE-Bench: Examining Post-Training Quantization for Mixture-of-Experts

Integrating Domain Knowledge into Large Language Models for Enhanced Fashion Recommendations