语言大模型 / LLM - arXivDaily 专题

2506.21972 2025-06-30 cs.CL cs.AI cs.CR cs.LG 85%

Advancing Jailbreak Strategies: A Hybrid Approach to Exploiting LLM Vulnerabilities and Bypassing Modern Defenses

Mohamed Ahmed, Mohamed Abdelmouty, Mingyu Kim, Gunvanth Kandula, Alex Park, James C. Davis

机构 * Purdue University（普渡大学）

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

2506.21901 2025-06-30 cs.DB 85%

A Survey of LLM Inference Systems

James Pan, Guoliang Li

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

Comments 25 pages

2506.21033 2025-06-27 cs.DC 85%

BLOCKS: Blockchain-supported Cross-Silo Knowledge Sharing for Efficient LLM Services

Zhaojiacheng Zhou, Hongze Liu, Shijing Yuan, Hanning Zhang, Jiong Lou, Chentao Wu, Jie Li

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

2503.13819 2025-06-26 cs.ET 85%

LLM-Empowered IoT for 6G Networks: Architecture, Challenges, and Solutions

Xiaopei Chen, Wen Wu, Liang Li, Fei Ji

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

Comments Accepted for publication in IEEE Internet of Things Magazine. \c{opyright} 2025 IEEE. The final version will be published in IEEE Xplore

2506.17865 2025-06-24 cs.CR cs.ET 85%

LASA: Enhancing SoC Security Verification with LLM-Aided Property Generation

Dinesh Reddy Ankireddy, Sudipta Paria, Aritra Dasgupta, Sandip Ray, Swarup Bhunia

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

Comments 9 pages, 5 figures, 5 tables

2506.16691 2025-06-23 cs.CV 85%

LaVi: Efficient Large Vision-Language Models via Internal Feature Modulation

Tongtian Yue, Longteng Guo, Yepeng Tang, Zijia Zhao, Xinxin Zhu, Hua Huang, Jing Liu

机构 * Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）； Institute of Information Science, Beijing Jiaotong University（北京交通大学信息科学研究院）； School of Artificial Intelligence, Beijing Normal University（北京师范大学人工智能学院）

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

2506.15947 2025-06-23 cs.NI eess.SP 85%

HybridRAG-based LLM Agents for Low-Carbon Optimization in Low-Altitude Economy Networks

Jinbo Wen, Cheng Su, Jiawen Kang, Jiangtian Nie, Yang Zhang, Jianhang Tang, Dusit Niyato, Chau Yuen

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

2503.13956 2025-06-19 cs.CV 85%

Improving LLM Video Understanding with 16 Frames Per Second

Yixuan Li, Changli Tang, Jimin Zhuang, Yudong Yang, Guangzhi Sun, Wei Li, Zejun Ma, Chao Zhang

机构 * Tsinghua University（清华大学）

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

2403.13198 2025-06-18 cs.RO 85%

LBAP: Improved Uncertainty Alignment of LLM Planners using Bayesian Inference

James F. Mullen, Dinesh Manocha

机构 * University of Maryland, College Park（马里兰大学学院公园分校）

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

2506.04015 2025-06-17 cs.IR 85%

GORACS: Group-level Optimal Transport-guided Coreset Selection for LLM-based Recommender Systems

Tiehua Mei, Hengrui Chen, Peng Yu, Jiaqing Liang, Deqing Yang

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

Comments Accepted by KDD 2025

2506.13494 2025-06-17 cs.CR 85%

Watermarking LLM-Generated Datasets in Downstream Tasks

Yugeng Liu, Tianshuo Cong, Michael Backes, Zheng Li, Yang Zhang

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

2506.12683 2025-06-17 cs.CV q-bio.QM 85%

Evaluating Cell Type Inference in Vision Language Models Under Varying Visual Context

Samarth Singhal, Sandeep Singhal

机构 * University of North Dakota（北达科他大学）

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

2506.12379 2025-06-17 cs.CL cs.AI cs.LG 85%

Training-free LLM Merging for Multi-task Learning

Zichuan Fu, Xian Wu, Yejing Wang, Wanyu Wang, Shanshan Ye, Hongzhi Yin, Yi Chang, Yefeng Zheng, Xiangyu Zhao

机构 * City University of Hong Kong（香港城市大学）； Tencent Jarvis Lab（腾讯 Jarvis 实验室）； University of Technology Sydney（悉尼大学）； University of Queensland（昆士兰大学）； Jilin University（吉林大学）； Westlake University（西湖大学）

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

Comments 14 pages, 6 figures

Journal ref ACL 2025 Main

2502.05234 2025-06-17 cs.LG cs.AI cs.CL 85%

Optimizing Temperature for Language Models with Multi-Sample Inference

Weihua Du, Yiming Yang, Sean Welleck

机构 * Language Technologies Institute, Carnegie Mellon University（卡内基梅隆大学语言技术研究所）

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

Comments ICML2025, 21 pages. Code available at https://github.com/StigLidu/TURN

2502.02013 2025-06-17 cs.LG cs.AI cs.CL 85%

Layer by Layer: Uncovering Hidden Representations in Language Models

Oscar Skean, Md Rifat Arefin, Dan Zhao, Niket Patel, Jalal Naghiyev, Yann LeCun, Ravid Shwartz-Ziv

机构 * University of Kentucky（肯塔基大学）； Mila-Quebec AI Institute（魁北克AI研究所）； University of Montreal（蒙特利尔大学）； New York University（纽约大学）； University of California, Los Angeles（加州大学洛杉矶分校）； Meta FAIR

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

Comments update for ICML2025 camera-ready

2506.11767 2025-06-16 cs.CY 85%

Designing Effective LLM-Assisted Interfaces for Curriculum Development

Abdolali Faraji, Mohammadreza Tavakoli, Mohammad Moein, Mohammadreza Molavi, Gábor Kismihók

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

Comments This is the preprint version of a paper accepted at AIED 2025. The final version will be published by Springer

2506.11141 2025-06-16 cs.SE cs.ET 85%

From over-reliance to smart integration: using Large-Language Models as translators between specialized modeling and simulation tools

Philippe J. Giabbanelli, John Beverley, Istvan David, Andreas Tolk

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

Comments Accepted at the Winter Simulation conference 2025, December, Seattle USA

2502.05202 2025-06-12 cs.CL cs.AI cs.LG 85%

Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies

Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Gaurav Jain, Oren Pereg, Moshe Wasserblat, David Harel

机构 * Weizmann Institute of Science（魏茨曼科学研究所）； Intel Labs（英特尔实验室）

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

Comments ICML'25 Oral (top %1)

2506.07627 2025-06-10 cs.CV 85%

Event-Priori-Based Vision-Language Model for Efficient Visual Understanding

Haotong Qin, Cheng Hu, Michele Magno

机构 * Center for Project-Based Learning D-ITET, ETH Zürich, Switzerland（项目基于学习中心D-ITET，苏黎世联邦理工学院，瑞士）； State Key Laboratory of Industrial Control Technology, Zhejiang University, China（工业控制技术国家重点实验室，浙江大学，中国）

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

2506.06579 2025-06-10 cs.LG cs.AI cs.CL cs.DC 85%

Towards Efficient Multi-LLM Inference: Characterization and Analysis of LLM Routing and Hierarchical Techniques

Adarsh Prasad Behera, Jaya Prakash Champati, Roberto Morabito, Sasu Tarkoma, James Gross

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

2506.06526 2025-06-10 eess.SP 85%

Prompting Wireless Networks: Reinforced In-Context Learning for Power Control

Hao Zhou, Chengming Hu, Dun Yuan, Ye Yuan, Di Wu, Xue Liu, Jianzhong, Zhang

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

Comments arXiv admin note: substantial text overlap with arXiv:2408.00214

2506.06343 2025-06-10 cs.CL cs.AI cs.LG cs.SD eess.AS 85%

TESU-LLM: Training Speech-LLMs Without Speech via Unified Encoder Alignment

Taesoo Kim, Jong Hwan Ko

机构 * KT Corporation（KT公司）； Department of Electrical and Computer Engineering（电气与计算机工程系）； Sungkyunkwan University（成均馆大学）

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

2501.18107 2025-06-10 cs.LG cs.AI cs.CL 85%

Scaling Inference-Efficient Language Models

Song Bian, Minghao Yan, Shivaram Venkataraman

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

Comments 21 pages, 18 figures, ICML 2025

2410.06511 2025-06-10 cs.CL cs.AI cs.DC cs.LG 85%

TorchTitan: One-stop PyTorch native solution for production ready LLM pre-training

Wanchao Liang, Tianyu Liu, Less Wright, Will Constable, Andrew Gu, Chien-Chin Huang, Iris Zhang, Wei Feng, Howard Huang, Junjie Wang, Sanket Purandare, Gokul Nadathur, Stratos Idreos

机构 * Meta ； Harvard University（哈佛大学）

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

2503.11315 2025-06-06 cs.CV cs.MM cs.SD eess.AS 85%

MMS-LLaMA: Efficient LLM-based Audio-Visual Speech Recognition with Minimal Multimodal Speech Tokens

Jeong Hun Yeo, Hyeongseop Rha, Se Jin Park, Yong Man Ro

机构 * Integrated Vision and Language Lab, KAIST, South Korea（集成视觉与语言实验室，韩国科学技术院）

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

Comments Accepted at Findings of ACL 2025. The code and models are available https://github.com/JeongHun0716/MMS-LLaMA

2502.00722 2025-06-06 cs.DC 85%

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs

Youhe Jiang, Fangcheng Fu, Xiaozhe Yao, Guoliang He, Xupeng Miao, Ana Klimovic, Bin Cui, Binhang Yuan, Eiko Yoneki

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

2506.03656 2025-06-05 cs.CR 85%

Client-Side Zero-Shot LLM Inference for Comprehensive In-Browser URL Analysis

Avihay Cohen

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

Comments 46 pages , 5 figures

2506.01967 2025-06-04 cs.LG cs.AI cs.CL 85%

Turning LLM Activations Quantization-Friendly

Patrik Czakó, Gábor Kertész, Sándor Szénási

机构 * Doctoral School of Applied Informatics and Applied Mathematics, Obuda University（应用信息与应用数学博士学院，奥布达大学）； John von Neumann Faculty of Informatics, Obuda University（约翰·冯·诺伊曼信息学院，奥布达大学）

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

Comments 6 pages, 5 figures. Accepted to SACI 2025 conference proceedings

2506.00888 2025-06-03 cs.SE 85%

An Integrated Platform for LEED Certification Automation Using Computer Vision and LLM-RAG

Jooyeol Lee

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

2502.17510 2025-06-02 cs.LG cs.AI cs.CL 85%

Recurrent Knowledge Identification and Fusion for Language Model Continual Learning

Yujie Feng, Xujia Wang, Zexin Lu, Shenghong Fu, Guangyuan Shi, Yongxin Xu, Yasha Wang, Philip S. Yu, Xu Chu, Xiao-Ming Wu

机构 * The Hong Kong Polytechnic University（香港理工大学）； Tsinghua University（清华大学）； Peking University（北京大学）； University of Illinois at Chicago（伊利诺伊大学香槟分校）

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

Comments Accepted to ACL 2025 (Main Conference)

1. 效率与部署 22368 篇

Advancing Jailbreak Strategies: A Hybrid Approach to Exploiting LLM Vulnerabilities and Bypassing Modern Defenses

A Survey of LLM Inference Systems

BLOCKS: Blockchain-supported Cross-Silo Knowledge Sharing for Efficient LLM Services

LLM-Empowered IoT for 6G Networks: Architecture, Challenges, and Solutions

LASA: Enhancing SoC Security Verification with LLM-Aided Property Generation

LaVi: Efficient Large Vision-Language Models via Internal Feature Modulation

HybridRAG-based LLM Agents for Low-Carbon Optimization in Low-Altitude Economy Networks

Improving LLM Video Understanding with 16 Frames Per Second

LBAP: Improved Uncertainty Alignment of LLM Planners using Bayesian Inference

GORACS: Group-level Optimal Transport-guided Coreset Selection for LLM-based Recommender Systems

Watermarking LLM-Generated Datasets in Downstream Tasks

Evaluating Cell Type Inference in Vision Language Models Under Varying Visual Context

Training-free LLM Merging for Multi-task Learning

Optimizing Temperature for Language Models with Multi-Sample Inference

Layer by Layer: Uncovering Hidden Representations in Language Models

Designing Effective LLM-Assisted Interfaces for Curriculum Development

From over-reliance to smart integration: using Large-Language Models as translators between specialized modeling and simulation tools

Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies

Event-Priori-Based Vision-Language Model for Efficient Visual Understanding

Towards Efficient Multi-LLM Inference: Characterization and Analysis of LLM Routing and Hierarchical Techniques

Prompting Wireless Networks: Reinforced In-Context Learning for Power Control

TESU-LLM: Training Speech-LLMs Without Speech via Unified Encoder Alignment

Scaling Inference-Efficient Language Models

TorchTitan: One-stop PyTorch native solution for production ready LLM pre-training

MMS-LLaMA: Efficient LLM-based Audio-Visual Speech Recognition with Minimal Multimodal Speech Tokens

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs

Client-Side Zero-Shot LLM Inference for Comprehensive In-Browser URL Analysis

Turning LLM Activations Quantization-Friendly

An Integrated Platform for LEED Certification Automation Using Computer Vision and LLM-RAG

Recurrent Knowledge Identification and Fusion for Language Model Continual Learning