语言大模型 / LLM - arXivDaily 专题

2509.14269 2025-09-23 cs.CL cs.AI 90%

SparseDoctor: Towards Efficient Chat Doctor with Mixture of Experts Enhanced Large Language Models

Jianbin Zhang, Yulin Zhu, Wai Lun Lo, Richard Tai-Chiu Hsung, Harris Sik-Ho Tsang, Kai Zhou

机构 * Department of Computer Science, Hong Kong Chu Hai College（香港 Chu Hai 学院计算机科学系）； Department of Computing, The Hong Kong Polytechnic University（香港理工大学计算机系）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.CL、cs.AI

2506.16792 2025-09-23 cs.CL cs.AI 90%

MIST: Jailbreaking Black-box Large Language Models via Iterative Semantic Tuning

Muyang Zheng, Yuanzhi Yao, Changting Lin, Caihong Kai, Yanxiang Chen, Zhiquan Liu

机构 * School of Computer Science and Information Engineering, Hefei University of Technology（合肥工业大学计算机科学与信息工程学院）； College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）； College of Cyber Security, Jinan University（暨南大学网络安全学院）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.CL、cs.AI

Comments 13 pages, 6 figures

2505.07004 2025-09-23 cs.LG 89%

GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidance

Jinuk Kim, Marwa El Halabi, Wonpyo Park, Clemens JS Schaefer, Deokjae Lee, Yeonhong Park, Jae W. Lee, Hyun Oh Song

机构 * Department of XXX, University of YYY, Location, Country ； School of ZZZ, Institute of WWW, Location, Country ； Department of Computer Science ； Engineering, Seoul National University ； Neural Processing Research Center ； Samsung AI Lab, Montreal ； Google

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；post-training(abstract)；分类 cs.LG

Comments ICML 2025

2505.17505 2025-09-23 cs.CL 89%

L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models

Xiaohao Liu, Xiaobo Xia, Weixiang Zhao, Manyi Zhang, Xianzhi Yu, Xiu Su, Shuo Yang, See-Kiong Ng, Tat-Seng Chua

机构 * National University of Singapore（新加坡国立大学）； Harbin Institute of Technology（哈尔滨工业大学）； Tsinghua University（清华大学）； Chinese Academy of Sciences（中国科学院）； Central South University（中南大学）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.CL

Comments Accepted by NeurIPS 2025. Codes are available at https://github.com/Xiaohao-Liu/L-MTP

2504.04653 2025-09-23 cs.CV cs.CL 89%

LEO-MINI: An Efficient Multimodal Large Language Model using Conditional Token Reduction and Mixture of Multi-Modal Experts

Yimu Wang, Mozhgan Nasr Azadani, Sean Sedwards, Krzysztof Czarnecki

机构 * University of Waterloo（滑铁卢大学）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.CL

Comments To appear at EMNLP 2025

2502.02818 2025-09-23 cs.DB cs.LG 89%

TranSQL+: Serving Large Language Models with SQL on Low-Resource Hardware

Wenbo Sun, Qiming Guo, Wenlu Wang, Rihan Hai

机构 * Delft University of Technology（代尔夫特理工大学）； Texas A\&M University - Corpus Christi（德克萨斯A&M大学-科罗萨尔）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.LG

Comments Accepted by SIGMOD2026

2509.13934 2025-09-23 eess.SY cs.LG cs.SY 89%

Large Language Model-Empowered Decision Transformer for UAV-Enabled Data Collection

Zhixion Chen, Jiangzhou Wang, Hyundong Shin, Arumugam Nallanathan

机构 * School of Electronic Engineering and Computer Science, Queen Mary University of London（伦敦女王学院电子工程与计算机科学学院）； National Mobile Communications Research Laboratory, Southeast University（东南大学国家移动通信研究中心）； Department of Electronics and Information Convergence Engineering, Kyung Hee University（韩国庆熙大学电子与信息融合工程系）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.LG

Comments 14pages, 8 figures

2410.21508 2025-09-23 cs.CL cs.AI 88%

Group-SAE: Efficient Training of Sparse Autoencoders for Large Language Models via Layer Groups

Davide Ghilardi, Federico Belotti, Marco Molinari, Tao Ma, Matteo Palmonari

机构 * University of Milan-Bicocca（米兰-比科卡大学）； London School of Economics（伦敦经济学院）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

Comments Accepted version at EMNLP'25

2509.16820 2025-09-23 cs.LG 88%

DISCO: Disentangled Communication Steering for Large Language Models

Max Torop, Aria Masoomi, Masih Eskandar, Jennifer Dy

机构 * Northeastern University（东北大学）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；分类 cs.LG

2509.16521 2025-09-23 cs.LG 88%

mmExpert: Integrating Large Language Models for Comprehensive mmWave Data Synthesis and Understanding

Yifan Yan, Shuai Yang, Xiuzhen Guo, Xiangguang Wang, Wei Chow, Yuanchao Shu, Shibo He

机构 * Zhejiang University（浙江大学）

专题命中效率与部署：large language model(title,abstract)；language model(title,abstract)；分类 cs.LG

Comments Accepted to ACM MobiHoc '25

2409.15520 2025-09-23 cs.LG cs.DC 87%

MobiZO: Enabling Efficient LLM Fine-Tuning at the Edge via Inference Engines

Lei Gao, Amir Ziashahabi, Yue Niu, Salman Avestimehr, Murali Annavaram

机构 * University of Southern California（南加州大学）

专题命中效率与部署：LLM(title,abstract)；large language model(abstract)；language model(abstract)；foundation model(abstract)

2506.09996 2025-09-23 cs.CL cs.CY 85%

From Judgment to Interference: Early Stopping LLM Harmful Outputs via Streaming Content Monitoring

Yang Li, Qiang Sheng, Yehan Yang, Xueyao Zhang, Juan Cao

机构 * Yang Li Media Synthesis and Forensics Lab, Institute of Computing Technology, Chinese Academy of Sciences University of Chinese Academy of Sciences（中国科学院计算技术研究所、中国科学院自动化研究所、中国科学院大学）； Qiang Sheng Media Synthesis and Forensics Lab, Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所、中国科学院自动化研究所）； Yehan Yang Media Synthesis and Forensics Lab, Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所、中国科学院自动化研究所）； Xueyao Zhang The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））； Juan Cao Media Synthesis and Forensics Lab, Institute of Computing Technology, Chinese Academy of Sciences University of Chinese Academy of Sciences（中国科学院计算技术研究所、中国科学院自动化研究所、中国科学院大学）

专题命中效率与部署：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

Comments NeurIPS 2025 Accepted Paper

2509.16443 2025-09-23 physics.app-ph cs.AI cs.PL 85%

LightCode: Compiling LLM Inference for Photonic-Electronic Systems

Ryan Tomich, Zhizhen Zhong, Dirk Englund

机构 * Research Laboratory of Electronics, Massachusetts Institute of Technology（麻省理工学院电子研究实验室）

专题命中效率与部署：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

Comments 9 pages, 8 figures

2509.16442 2025-09-23 cs.IR cs.CL 85%

Evaluating the Effectiveness and Scalability of LLM-Based Data Augmentation for Retrieval

Pranjal A. Chitale, Bishal Santra, Yashoteja Prabhu, Amit Sharma

机构 * Microsoft Research India（微软研究院印度分部）

专题命中效率与部署：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

Comments EMNLP 2025 (MAIN Conference)

2503.01606 2025-09-23 cs.CL cs.AI cs.CY cs.LG 85%

Beyond Prompting: An Efficient Embedding Framework for Open-Domain Question Answering

Zhanghao Hu, Hanqi Yan, Qinglin Zhu, Zhenyi Shen, Yulan He, Lin Gui

机构 * King’s College London（伦敦国王学院）； The Alan Turing Institute（艾伦·图灵研究所）

专题命中效率与部署：prompting(title)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted in ACL 2025 Main, Project link: https://zhanghao-acl25-embqa.github.io/ACL2025-EmbQA/

2509.16597 2025-09-23 cs.CL 84%

MCP: A Control-Theoretic Orchestration Framework for Synergistic Efficiency and Interpretability in Multimodal Large Language Models

Luyan Zhang

机构 * Northeastern University（东北大学）

专题命中效率与部署：large language model(title)；language model(title)；分类 cs.CL

Comments 13 pages, 6 figures, 2 tables

2502.00085 2025-09-23 cs.CL 84%

Efficient Beam Search for Large Language Models Using Trie-Based Decoding

Brian J Chan, MaoXun Huang, Jui-Hung Cheng, Chao-Ting Chen, Hen-Hsen Huang

机构 * Department of Computer Science, National Chengchi University（国立成功大学计算机科学系）； Department of Computer Science, Cornell University（康奈尔大学计算机科学系）； Institute of Information Science, Academia Sinica（中华科学研究院资讯研究所）

专题命中效率与部署：large language model(title)；language model(title)；分类 cs.CL

Comments 13 pages, accepted as a main conference paper at EMNLP 2025

2509.17096 2025-09-23 cs.SE cs.AI cs.HC 83%

Prompt-with-Me: in-IDE Structured Prompt Management for LLM-Driven Software Engineering

Ziyou Li, Agnia Sergeyuk, Maliheh Izadi

机构 * Delft University of Technology（代尔夫特理工大学）； JetBrains Research（JetBrains研究）

专题命中效率与部署：LLM(title)；large language model(abstract)；language model(abstract)；分类 cs.AI

Comments Accepted in the 40th IEEE/ACM International Conference on Automated Software Engineering, ASE 2025 (Industry track)

2509.11815 2025-09-23 cs.CV cs.AI 83%

SpecVLM: Fast Speculative Decoding in Vision-Language Models

Haiduo Huang, Fuwei Yang, Zhenhua Liu, Xuanwu Yin, Dong Li, Pengju Ren, Emad Barsoum

机构 * Advanced Micro Devices, Inc.（先进微器件公司）； Institute of Artificial Intelligence and Robotics（人工智能与机器人研究院）

专题命中效率与部署：language model(title,abstract)；large language model(abstract)；分类 cs.AI

2405.15269 2025-09-23 cs.CV cs.LG 83%

Test-Time Multimodal Backdoor Detection by Contrastive Prompting

Yuwei Niu, Shuo He, Qi Wei, Zongyu Wu, Feng Liu, Lei Feng

机构 * Chongqing University（重庆大学）； Nanyang Technological University（南洋理工大学）； Penn State University（宾夕法尼亚州立大学）； University of Melbourne（墨尔本大学）； Southeast University（东南大学）

专题命中效率与部署：prompting(title,abstract)；language model(abstract)；分类 cs.LG

Comments Accepted to ICML2025

2509.16995 2025-09-23 cs.DC 82%

MoA-Off: Adaptive Heterogeneous Modality-Aware Offloading with Edge-Cloud Collaboration for Efficient Multimodal LLM Inference

Zheming Yang, Qi Guo, Yunqing Hu, Chang Zhao, Chang Zhang, Jian Zhao, Wen Ji

专题命中效率与部署：LLM(title)；large language model(abstract)；language model(abstract)

Comments 5 pages, 4 figures

2509.16278 2025-09-23 cs.CL cs.LG 81%

Language Modeling with Learned Meta-Tokens

Alok N. Shah, Khush Gupta, Keshav Ramji, Pratik Chaudhari

机构 * University of Pennsylvania（宾夕法尼亚大学）； IBM Research AI（IBM人工智能研究）

专题命中效率与部署：language model(title,abstract)；分类 cs.CL、cs.LG

2509.17730 2025-09-23 cs.LG cs.CL 79%

ConfClip: Confidence-Weighted and Clipped Reward for Reinforcement Learning in LLMs

Bonan Zhang, Zhongqi Chen, Bowen Song, Qinya Li, Fan Wu, Guihai Chen

机构 * Shanghai Jiao Tong University Ant Group（上海交通大学-蚂蚁集团）

专题命中效率与部署：large language model(abstract)；language model(abstract)；instruction tuning(abstract)；分类 cs.CL、cs.LG

2509.17553 2025-09-23 cs.AI cs.DB cs.LG 79%

MontePrep: Monte-Carlo-Driven Automatic Data Preparation without Target Data Instances

Congcong Ge, Yachuan Liu, Yixuan Tang, Yifan Zhu, Yaofeng Tu, Yunjun Gao

机构 * Zhejiang University（浙江大学）； National University of Singapore（新加坡国立大学）； ZTE Corporation（中兴通讯公司）

专题命中效率与部署：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

2509.17357 2025-09-23 cs.DC 78%

Cronus: Efficient LLM inference on Heterogeneous GPU Clusters via Partially Disaggregated Prefill

Yunzhao Liu, Qiang Xu, Y. Charlie Hu

专题命中效率与部署：LLM(title,abstract)

2509.17486 2025-09-23 cs.CL 77%

AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation

Lvzhou Luo, Yixuan Cao, Ping Luo

机构 * Key Lab of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences (CAS)（智能信息处理重点实验室，计算技术研究所，中国科学院）； State Key Lab of AI Safety（人工智能安全国家重点实验室）； University of Chinese Academy of Sciences, CAS（中国科学院大学）

专题命中效率与部署：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

Comments Accepted at EMNLP 2025 (Findings)

2509.16686 2025-09-23 cs.CL 77%

EG-MLA: Embedding-Gated Multi-head Latent Attention for Scalable and Efficient LLMs

Zhengge Cai, Haowen Hou

专题命中效率与部署：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

2509.16649 2025-09-23 cs.SD cs.AI eess.AS 77%

AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval

Hyun Jun Kim, Hyeong Yong Choi, Changwon Lim

机构 * AISTAT LAB（AISTAT实验室）

专题命中效率与部署：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

Comments 5 pages, 1 figure, DCASE2025 Task2 technical report

2508.01332 2025-09-23 cs.CR cs.AI 77%

BlockA2A: Towards Secure and Verifiable Agent-to-Agent Interoperability

Zhenhua Zou, Zhuotao Liu, Lepeng Zhao, Qiuyang Zhan

机构 * Tsinghua University（清华大学）

专题命中效率与部署：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

Comments 43 pages

2509.16870 2025-09-23 cs.SE cs.CR 75%

DecipherGuard: Understanding and Deciphering Jailbreak Prompts for a Safer Deployment of Intelligent Software Systems

Rui Yang, Michael Fu, Chakkrit Tantithamthavorn, Chetan Arora, Gunel Gulmammadova, Joey Chua

专题命中效率与部署：LLM(abstract)；large language model(abstract)；language model(abstract)

Comments Under Review

1. 效率与部署 47 篇

SparseDoctor: Towards Efficient Chat Doctor with Mixture of Experts Enhanced Large Language Models

MIST: Jailbreaking Black-box Large Language Models via Iterative Semantic Tuning

GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidance

L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models

LEO-MINI: An Efficient Multimodal Large Language Model using Conditional Token Reduction and Mixture of Multi-Modal Experts

TranSQL+: Serving Large Language Models with SQL on Low-Resource Hardware

Large Language Model-Empowered Decision Transformer for UAV-Enabled Data Collection

Group-SAE: Efficient Training of Sparse Autoencoders for Large Language Models via Layer Groups

DISCO: Disentangled Communication Steering for Large Language Models

mmExpert: Integrating Large Language Models for Comprehensive mmWave Data Synthesis and Understanding

MobiZO: Enabling Efficient LLM Fine-Tuning at the Edge via Inference Engines

From Judgment to Interference: Early Stopping LLM Harmful Outputs via Streaming Content Monitoring

LightCode: Compiling LLM Inference for Photonic-Electronic Systems

Evaluating the Effectiveness and Scalability of LLM-Based Data Augmentation for Retrieval

Beyond Prompting: An Efficient Embedding Framework for Open-Domain Question Answering

MCP: A Control-Theoretic Orchestration Framework for Synergistic Efficiency and Interpretability in Multimodal Large Language Models

Efficient Beam Search for Large Language Models Using Trie-Based Decoding

Prompt-with-Me: in-IDE Structured Prompt Management for LLM-Driven Software Engineering

SpecVLM: Fast Speculative Decoding in Vision-Language Models

Test-Time Multimodal Backdoor Detection by Contrastive Prompting

MoA-Off: Adaptive Heterogeneous Modality-Aware Offloading with Edge-Cloud Collaboration for Efficient Multimodal LLM Inference

Language Modeling with Learned Meta-Tokens

ConfClip: Confidence-Weighted and Clipped Reward for Reinforcement Learning in LLMs

MontePrep: Monte-Carlo-Driven Automatic Data Preparation without Target Data Instances

Cronus: Efficient LLM inference on Heterogeneous GPU Clusters via Partially Disaggregated Prefill

AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation

EG-MLA: Embedding-Gated Multi-head Latent Attention for Scalable and Efficient LLMs

AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval

BlockA2A: Towards Secure and Verifiable Agent-to-Agent Interoperability

DecipherGuard: Understanding and Deciphering Jailbreak Prompts for a Safer Deployment of Intelligent Software Systems