语言大模型 / LLM - arXivDaily 专题

2511.18589 2025-11-25 cs.SE cs.AI 85%

Strategic Decision Framework for Enterprise LLM Adoption

企业LLM采纳的战略决策框架

Michael Trusov, Minha Hwang, Zainab Jamal, Swarup Chandra

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

AI总结本文提出一个六步决策框架，帮助企业安全高效地采用LLM，平衡技术能力与业务目标。

Comments 14 pages, 1 key figure

2505.16570 2025-11-25 cs.CL 85%

URLs Help, Topics Guide: Understanding Metadata Utility in LLM Training

URL助益，主题引导：理解元数据在LLM训练中的效用

Dongyang Fan, Vinko Sabolčec, Martin Jaggi

机构 * EPFL（苏黎世联邦理工学院）

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

AI总结本研究探讨了元数据在LLM训练中的效用，发现URL上下文能提升训练效率，而主题和格式信息则能引导生成输出，但不加速训练。

Comments NeurIPS 2025, Camera Ready

2511.17793 2025-11-25 cs.CV cs.LG 85%

Attention Guided Alignment in Efficient Vision-Language Models

注意力引导的高效视觉-语言模型

Shweta Mahajan, Hoang Le, Hyojin Park, Farzad Farhadzadeh, Munawar Hayat, Fatih Porikli

机构 * Qualcomm AI Research（高通人工智能研究）

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

AI总结本文提出AGE-VLM，通过交错交叉注意力层和空间知识提取，减少高效视觉-语言模型中的幻觉问题。

Comments 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop on Efficient Reasoning

2511.17683 2025-11-25 cs.CY cs.AI 85%

Datacenters in the Desert: Feasibility and Sustainability of LLM Inference in the Middle East

沙漠中的数据中心：中东地区大型语言模型推理的可行性与可持续性

Lara Hassan, Mohamed ElZeftawy, Abdulrahman Mahmoud

机构 * Department of Computer Science（计算机科学系）； MBZUAI（马斯克大学人工智能研究所）

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

AI总结本文研究了在中东沙漠地区部署大型语言模型推理数据中心的可行性与可持续性，通过实证分析能耗和碳足迹，评估不同地理区域的气候意识AI部署策略。

Comments 3 pages, 1 figure

Journal ref DCEE-2025@ISCA-2025

2511.15996 2025-11-21 cs.IR cs.CL 85%

QueryGym: A Toolkit for Reproducible LLM-Based Query Reformulation

QueryGym: 一个用于可重复LLM查询重述的工具包

Amin Bigdeli, Radin Hamidi Rad, Mert Incesu, Negar Arabzadeh, Charles L. A. Clarke, Ebrahim Bagheri

机构 * University of Waterloo（滑铁卢大学）； Mila – Quebec AI Institute（魁北克人工智能研究所）； University of Toronto（多伦多大学）； University of California, Berkeley（加州大学伯克利分校）

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

AI总结 QueryGym是一个用于可重复LLM查询重述的开源工具包，提供统一框架和多种集成支持，促进方法比较与基准测试。

Comments 4 pages

2511.11729 2025-11-20 cs.DC cs.LG 85%

Harli: SLO-Aware Co-location of LLM Inference and PEFT-based Finetuning on Model-as-a-Service Platforms

Ao Xu, Han Zhao, Weihao Cui, Quan Chen, Yukang Chen, Shulai Zhang, Shuang Chen, Jiemin Jiang, Zhibin Yu, Minyi Guo

机构 * Shanghai Jiao Tong University（上海交通大学）； Zhejiang University（浙江大学）； Chinese Academy of Science（中国科学院）

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

2511.14124 2025-11-19 cs.DC cs.LG 85%

10Cache: Heterogeneous Resource-Aware Tensor Caching and Migration for LLM Training

Sabiha Afroz, Redwan Ibne Seraj Khan, Hadeel Albahar, Jingoo Han, Ali R. Butt

机构 * Kuwait University, Kuwait（科威特大学）

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

Comments This paper accepted for presentation to the 16th ACM Symposium on Cloud Computing (SOCC'25)

2511.12376 2025-11-19 cs.LG 85%

BitSnap: Checkpoint Sparsification and Quantization in LLM Training

Yanxin Peng, Qingping Li, Baodong Wu, Shigang Li, Guohao Dai, Shengen Yan, Yu Wang

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

Comments 12 pages, numerous figures

2511.09087 2025-11-19 cs.NI cs.AI 85%

Tele-LLM-Hub: Building Context-Aware Multi-Agent LLM Systems for Telecom Networks

Pranshav Gajjar, Cong Shen, Vijay K Shah

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

2502.19662 2025-11-18 cs.AR cs.AI 85%

HALO: Hardware-aware quantization with low critical-path-delay weights for LLM acceleration

Rohan Juneja, Shivam Aggarwal, Safeen Huda, Tulika Mitra, Li-Shiuan Peh

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

2511.12217 2025-11-18 cs.LG 85%

AlignTree: Efficient Defense Against LLM Jailbreak Attacks

Gil Goren, Shahar Katz, Lior Wolf

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

Comments Accepted as an Oral Presentation at the 40th AAAI Conference on Artificial Intelligence (AAAI-26), January 2026

2511.12031 2025-11-18 cs.DC cs.AI 85%

Striking the Right Balance between Compute and Copy: Improving LLM Inferencing Under Speculative Decoding

Arun Ramachandran, Ramaswamy Govindarajan, Murali Annavaram, Prakash Raghavendra, Hossein Entezari Zarch, Lei Gao, Chaoyi Jiang

机构 * Indian Institute of Science (IISc)（印度科学研究所）

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

2511.11788 2025-11-18 cs.MA cs.AI 85%

MALBO: Optimizing LLM-Based Multi-Agent Teams via Multi-Objective Bayesian Optimization

Antonio Sabbatella

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

Comments Master's Thesis, University of Milano-Bicocca, 2025

2511.11733 2025-11-18 cs.DC cs.AI 85%

Speculative Decoding in Decentralized LLM Inference: Turning Communication Latency into Computation Throughput

Jingwei Song, Wanyi Chen, Xinyuan Song, Max, Chris Tong, Gufeng Chen, Tianyi Zhao, Eric Yang, Bill Shi, Lynn Ai

机构 * Department of XXX, University of YYY, Location, Country（XXX系，YYY大学，地点，国家）； School of ZZZ, Institute of WWW, Location, Country（ZZZ学院，WWW研究所，地点，国家）

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

Comments 6 pages, 2 figures, 2 tables. Uses ICML 2025 style

2511.11621 2025-11-18 cs.DC cs.AI cs.NI 85%

AIvailable: A Software-Defined Architecture for LLM-as-a-Service on Heterogeneous and Legacy GPUs

Pedro Antunes, Ana Rita Ortigoso, Gabriel Vieira, Daniel Fuentes, Luís Frazão, Nuno Costa, António Pereira

机构 * Computer Science and Communication Research Centre（计算机科学与通信研究中心）； Polytechnic University of Leiria（利尔亚理工学院）

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

2511.11574 2025-11-18 cs.LG 85%

LLM on a Budget: Active Knowledge Distillation for Efficient Classification of Large Text Corpora

Viviana Luccioli, Rithika Iyengar, Ryan Panley, Flora Haberkorn, Xiaoyu Ge, Leland Crane, Nitish Sinha, Seung Jung Lee

机构 * Board of Governors of the Federal Reserve System（联邦储备系统委员会）

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

2511.09820 2025-11-14 cs.CV cs.AI 85%

From Street to Orbit: Training-Free Cross-View Retrieval via Location Semantics and LLM Guidance

Jeongho Min, Dongyoung Kim, Jaehyup Lee

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

Comments Accepted to WACV 2026, 10pages, 4 figures

2511.09025 2025-11-13 cs.LG 85%

FLAD: Federated Learning for LLM-based Autonomous Driving in Vehicle-Edge-Cloud Networks

Tianao Xiang, Mingjian Zhi, Yuanguo Bi, Lin Cai, Yuhao Chen

机构 * Northeastern University（东北大学）； University of Victoria（维多利亚大学）

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

2511.07505 2025-11-12 cs.CR cs.AI 85%

FedRW: Efficient Privacy-Preserving Data Reweighting for Enhancing Federated Learning of Language Models

Pukang Ye, Junwei Luo, Xiaolei Dong, Yunbo Yang

机构 * East China Normal University（华东师范大学）； Wuhan University（武汉大学）； Zhejiang University（浙江大学）； ZStack ； Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security Q（杭州高新区（滨江）区块链与数据安全研究院）

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

Comments Accepted at NeurIPS 2025. Code is available at https://github.com/Hecateto/FedRW

2511.07427 2025-11-12 cs.DC cs.AI 85%

DynaKV: Enabling Accurate and Efficient Long-Sequence LLM Decoding on Smartphones

Tuowei Wang, Minxing Huang, Fengzu Li, Ligeng Chen, Jinrui Zhang, Ju Ren

机构 * Tsinghua University（清华大学）； Honor Device Co., Ltd.（荣耀设备有限公司）

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

2509.08736 2025-11-11 cs.LG 85%

ChemBOMAS: Accelerated BO in Chemistry with LLM-Enhanced Multi-Agent System

Dong Han, Zhehong Ai, Pengxiang Cai, Shanya Lu, Jianpeng Chen, Zihao Ye, Shuzhou Sun, Ben Gao, Lingli Ge, Weida Wang, Xiangxin Zhou, Xihui Liu, Mao Su, Wanli Ouyang, Lei Bai, Dongzhan Zhou, Tao Xu, Yuqiang Li, Shufei Zhang

机构 * Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）； Tongji University（同济大学）； The University of Hong Kong（香港大学）； Fudan University（复旦大学）

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

2409.13949 2025-11-11 cs.CL 85%

Mufu: Multilingual Fused Learning for Low-Resource Translation with LLM

Zheng Wei Lim, Nitish Gupta, Honglin Yu, Trevor Cohn

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

Comments 29 pages

2511.06262 2025-11-11 cs.AI cs.CY 85%

GAIA: A General Agency Interaction Architecture for LLM-Human B2B Negotiation & Screening

Siming Zhao, Qi Li

机构 * Alibaba.com US E-Commerce（阿里巴巴美国电子商务）； Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））

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

2511.05502 2025-11-11 cs.AR cs.AI 85%

Production-Grade Local LLM Inference on Apple Silicon: A Comparative Study of MLX, MLC-LLM, Ollama, llama.cpp, and PyTorch MPS

Varun Rajesh, Om Jodhpurkar, Pooja Anbuselvan, Mantinder Singh, Ashok Jallepali, Shantanu Godbole, Pradeep Kumar Sharma, Hritvik Shrivastava

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

2511.01602 2025-11-10 cs.DB cs.LG 85%

L2T-Tune:LLM-Guided Hybrid Database Tuning with LHS and TD3

Xinyue Yang, Chen Zheng, Yaoyang Hou, Renhao Zhang, Yinyan Zhang, Yanjun Wu, Heng Zhang

机构 * University of Chinese Academy of Sciences（中国科学院大学）； Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）； Hangzhou Institute for Advanced Study, UCAS（杭州高等研究院）

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

2504.10816 2025-11-10 cs.IR cs.CL 85%

CSPLADE: Learned Sparse Retrieval with Causal Language Models

Zhichao Xu, Aosong Feng, Yijun Tian, Haibo Ding, Lin Lee Cheong

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

Comments IJCNLP-AACL 2025 Main

2511.02755 2025-11-05 cs.CL 85%

Controlling Performance and Budget of a Centralized Multi-agent LLM System with Reinforcement Learning

Bowen Jin, TJ Collins, Donghan Yu, Mert Cemri, Shenao Zhang, Mengyu Li, Jay Tang, Tian Qin, Zhiyang Xu, Jiarui Lu, Guoli Yin, Jiawei Han, Zirui Wang

机构 * University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）； Apple（苹果公司）； University of California, Berkeley（加州大学伯克利分校）； Northwestern University（西北大学）； Harvard University（哈佛大学）； Virginia Tech（弗吉尼亚理工大学）

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

Comments 14 pages

2511.01866 2025-11-05 cs.DC cs.AI cs.AR 85%

EdgeReasoning: Characterizing Reasoning LLM Deployment on Edge GPUs

Benjamin Kubwimana, Qijing Huang

机构 * NVIDIA

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

Comments Published in the Proceedings of the 2025 IEEE International Symposium on Workload Characterization (IISWC 2025)

2511.01850 2025-11-04 cs.SE cs.AI 85%

SmartMLOps Studio: Design of an LLM-Integrated IDE with Automated MLOps Pipelines for Model Development and Monitoring

Jiawei Jin, Yingxin Su, Xiaotong Zhu

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

2511.00096 2025-11-04 cs.MA cs.AI cs.CY 85%

Urban-MAS: Human-Centered Urban Prediction with LLM-Based Multi-Agent System

Shangyu Lou

机构 * University of California, Santa Barbara \& San Diego State University California USA ； University of California, Santa Barbara \& San Diego State University

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

Comments Accepted to The 3rd ACM SIGSPATIAL International Workshop on Advances in Urban AI (UrbanAI'25)

1. 效率与部署 22313 篇

Strategic Decision Framework for Enterprise LLM Adoption

URLs Help, Topics Guide: Understanding Metadata Utility in LLM Training

Attention Guided Alignment in Efficient Vision-Language Models

Datacenters in the Desert: Feasibility and Sustainability of LLM Inference in the Middle East

QueryGym: A Toolkit for Reproducible LLM-Based Query Reformulation

Harli: SLO-Aware Co-location of LLM Inference and PEFT-based Finetuning on Model-as-a-Service Platforms

10Cache: Heterogeneous Resource-Aware Tensor Caching and Migration for LLM Training

BitSnap: Checkpoint Sparsification and Quantization in LLM Training

Tele-LLM-Hub: Building Context-Aware Multi-Agent LLM Systems for Telecom Networks

HALO: Hardware-aware quantization with low critical-path-delay weights for LLM acceleration

AlignTree: Efficient Defense Against LLM Jailbreak Attacks

Striking the Right Balance between Compute and Copy: Improving LLM Inferencing Under Speculative Decoding

MALBO: Optimizing LLM-Based Multi-Agent Teams via Multi-Objective Bayesian Optimization

Speculative Decoding in Decentralized LLM Inference: Turning Communication Latency into Computation Throughput

AIvailable: A Software-Defined Architecture for LLM-as-a-Service on Heterogeneous and Legacy GPUs

LLM on a Budget: Active Knowledge Distillation for Efficient Classification of Large Text Corpora

From Street to Orbit: Training-Free Cross-View Retrieval via Location Semantics and LLM Guidance

FLAD: Federated Learning for LLM-based Autonomous Driving in Vehicle-Edge-Cloud Networks

FedRW: Efficient Privacy-Preserving Data Reweighting for Enhancing Federated Learning of Language Models

DynaKV: Enabling Accurate and Efficient Long-Sequence LLM Decoding on Smartphones

ChemBOMAS: Accelerated BO in Chemistry with LLM-Enhanced Multi-Agent System

Mufu: Multilingual Fused Learning for Low-Resource Translation with LLM

GAIA: A General Agency Interaction Architecture for LLM-Human B2B Negotiation & Screening

Production-Grade Local LLM Inference on Apple Silicon: A Comparative Study of MLX, MLC-LLM, Ollama, llama.cpp, and PyTorch MPS

L2T-Tune:LLM-Guided Hybrid Database Tuning with LHS and TD3

CSPLADE: Learned Sparse Retrieval with Causal Language Models

Controlling Performance and Budget of a Centralized Multi-agent LLM System with Reinforcement Learning

EdgeReasoning: Characterizing Reasoning LLM Deployment on Edge GPUs

SmartMLOps Studio: Design of an LLM-Integrated IDE with Automated MLOps Pipelines for Model Development and Monitoring

Urban-MAS: Human-Centered Urban Prediction with LLM-Based Multi-Agent System