语言大模型 / LLM - arXivDaily 专题

2511.01695 2025-12-01 cs.LG eess.SP 89%

Collaborative Large Language Model Inference via Resource-Aware Parallel Speculative Decoding

通过资源感知的并行推测解码实现协作式大语言模型推理

Jungyeon Koh, Hyun Jong Yang

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

AI总结本文提出一种统一框架，通过多智能体深度强化学习优化用户关联和资源分配，实现高效的并行推测解码，减少端到端延迟28%以上，提升移动边缘计算中大语言模型服务的性能。

2505.01658 2025-11-27 cs.CL 89%

A Survey on Inference Engines for Large Language Models: Perspectives on Optimization and Efficiency

对大型语言模型推理引擎的综述：优化与效率的视角

Sihyeong Park, Sungryeol Jeon, Chaelyn Lee, Seokhun Jeon, Byung-Soo Kim, Jemin Lee

机构 * Korea Electronics Technology Institute（韩国电子技术研究所）； Electronics and Telecommunications Research Institute（电子电信研究院）

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

AI总结本文综述了大型语言模型推理引擎的优化与效率，评估了25种开源和商用引擎，探讨了其设计目标及生态系统成熟度，并提供未来研究方向和公开存储库。

Comments Under review; 106 pages; 46 figures

2503.08524 2025-11-27 cs.CL 89%

Position-Aware Depth Decay Decoding ($D^3$): Boosting Large Language Model Inference Efficiency

位置感知深度衰减解码（D³）：提升大语言模型推理效率

Siqi Fan, Xuezhi Fang, Xingrun Xing, Peng Han, Shuo Shang, Yequan Wang

机构 * University of Electronic Science and Technology of China（电子科技大学）； Beijing Academy of Artificial Intelligence（北京人工智能研究院）； Institute of Computing Automation, Chinese Academy of Sciences（中国科学院计算技术研究所）

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

AI总结 D³通过位置感知深度衰减解码提升大语言模型推理效率，实现无需重新训练的高效生成。

2511.17205 2025-11-24 cs.CL 89%

E$^3$-Pruner: Towards Efficient, Economical, and Effective Layer Pruning for Large Language Models

E$^3$-Pruner：面向大语言模型高效、经济和有效的层剪枝

Tao Yuan, Haoli Bai, Yinfei Pan, Xuyang Cao, Tianyu Zhang, Lu Hou, Ting Hu, Xianzhi Yu

机构 * Huawei Technologies（华为技术有限公司）； Tsinghua Shenzhen International Graduate School（清华大学深圳国际研究生学院）

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

AI总结 E$^3$-Pruner通过可微掩码优化和熵感知知识蒸馏，实现了大语言模型的高效、经济和有效层剪枝，显著提升性能与效率。

2511.15992 2025-11-21 cs.AI 89%

Detecting Sleeper Agents in Large Language Models via Semantic Drift Analysis

通过语义漂移分析检测大语言模型中的休眠代理

Shahin Zanbaghi, Ryan Rostampour, Farhan Abid, Salim Al Jarmakani

机构 * School of Computer Science, University of Windsor（计算机科学学院，温莎大学）

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

AI总结通过语义漂移分析与canary基线比较，实现大语言模型中休眠代理的实时检测，准确率达92.5%。

Comments 7 pages, 3 figures, 1 table

2511.14650 2025-11-19 cs.AI 89%

AutoTool: Efficient Tool Selection for Large Language Model Agents

Jingyi Jia, Qinbin Li

机构 * Jingyi Jia, Qinbin Li

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

Comments Accepted by AAAI 2026, 18 pages, 11 figures, Code: https://github.com/jiajingyyyyyy/AutoTool

2309.06706 2025-11-18 cs.CL 89%

Simultaneous Machine Translation with Large Language Models

Minghan Wang, Jinming Zhao, Thuy-Trang Vu, Fatemeh Shiri, Ehsan Shareghi, Gholamreza Haffari

机构 * Department of Data Science & AI, Monash University（数据科学与人工智能系，墨尔本大学）

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 to ALTA 2024

2511.09231 2025-11-14 cs.SE cs.AI 89%

Leveraging Large Language Models for Use Case Model Generation from Software Requirements

Tobias Eisenreich, Nicholas Friedlaender, Stefan Wagner

机构 * School of Computation, Information and Technology（计算、信息与技术学院）； Technical University of Munich（慕尼黑技术大学）

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

Comments Accepted at the Intelligent Software Engineering Workshop (ISE 2025) at ASE 2025

2511.09880 2025-11-14 cs.CL cs.CR 89%

EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language Models

Jialin Wu, Kecen Li, Zhicong Huang, Xinfeng Li, Xiaofeng Wang, Cheng Hong

机构 * Ant Group（蚂蚁集团）； Nanyang Technological University（南洋理工大学）

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 IEEE Symposium on Security and Privacy (S&P) 2026

2508.01365 2025-11-12 cs.CR cs.CL 89%

ConfGuard: A Simple and Effective Backdoor Detection for Large Language Models

Zihan Wang, Rui Zhang, Hongwei Li, Wenshu Fan, Wenbo Jiang, Qingchuan Zhao, Guowen Xu

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 This is an extended version of the copyrighted publication at AAAI

2503.09205 2025-11-12 cs.MM cs.CL cs.IR cs.SD eess.AS 89%

Quality Over Quantity? LLM-Based Curation for a Data-Efficient Audio-Video Foundation Model

Ali Vosoughi, Dimitra Emmanouilidou, Hannes Gamper

机构 * University of Rochester（罗切斯特大学）； Microsoft Research（微软研究院）

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

Comments Accepted at EUSIPCO 2025 - 5 pages, 5 figures, 2 tables

2502.16002 2025-11-11 cs.CL 89%

KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse

Jingbo Yang, Bairu Hou, Wei Wei, Yujia Bao, Shiyu Chang

机构 * Department of Computer Science UC Santa Barbara（计算机科学系 UC 圣塔芭芭拉大学）； Center for Advanced AI Accenture（先进人工智能中心江苏 Accenture）

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

2511.00457 2025-11-11 cs.AI 89%

GraphChain: Large Language Models for Large-scale Graph Analysis via Tool Chaining

Chunyu Wei, Wenji Hu, Xingjia Hao, Xin Wang, Yifan Yang, Yueguo Chen, Yang Tian, Yunhai Wang

机构 * Renmin University of China（中国人民大学）； Guangxi University（广西大学）； Beijing Jiaotong University（北京交通大学）

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

Comments Accepted by NeurIPS 2025

2508.09091 2025-11-11 cs.CL 89%

Utilizing Multilingual Encoders to Improve Large Language Models for Low-Resource Languages

Imalsha Puranegedara, Themira Chathumina, Nisal Ranathunga, Nisansa de Silva, Surangika Ranathunga, Mokanarangan Thayaparan

机构 * School of Mathematical and Computational Sciences, Massey University, Auckland, New Zealand（数学与计算科学学院，梅西大学，奥克兰，新西兰）； Department of Computer Science, The Open University, Milton Keynes, United Kingdom（计算机科学系，开放大学，米利顿凯斯，英国）

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

2501.16309 2025-11-11 physics.med-ph cs.AI 89%

Evaluating The Performance of Using Large Language Models to Automate Summarization of CT Simulation Orders in Radiation Oncology

Meiyun Cao, Shaw Hu, Jason Sharp, Edward Clouser, Jason Holmes, Linda L. Lam, Xiaoning Ding, Diego Santos Toesca, Wendy S. Lindholm, Samir H. Patel, Sujay A. Vora, Peilong Wang, Wei Liu

机构 * Department of Radiation Oncology, Mayo Clinic, Phoenix, AZ 85054（放射肿瘤科，梅奥诊所，凤凰城，AZ 85054）

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

2511.03137 2025-11-06 cs.AI 89%

Using Multi-modal Large Language Model to Boost Fireworks Algorithm's Ability in Settling Challenging Optimization Tasks

Shipeng Cen, Ying Tan

机构 * School of Intelligence Science ； Technology, Institute for Artificial Intellignce, Peking University, Beijing, China ； Technology, Institute for Artificial Intellignce, National Key Laboratory of General Artificial Intelligence, Peking University, Beijing, China

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

2506.19248 2025-11-06 cs.LG 89%

Inference-Time Reward Hacking in Large Language Models

Hadi Khalaf, Claudio Mayrink Verdun, Alex Oesterling, Himabindu Lakkaraju, Flavio du Pin Calmon

机构 * Harvard University（哈佛大学）

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

Comments Accepted to NeurIPS 2025 (Spotlight Paper)

2510.26446 2025-10-31 cs.CL 89%

1+1>2: A Synergistic Sparse and Low-Rank Compression Method for Large Language Models

Zeliang Zong, Kai Zhang, Zheyang Li, Wenming Tan, Ye Ren, Yiyan Zhai, Jilin Hu

机构 * Hikvision Research Institute（海康威视研究院）； School of Data Science and Engineering, East China Normal University（华东师范大学数据科学与工程学院）

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

Comments 15 pages, 6 figures, EMNLP 2025 findings

2510.20428 2025-10-24 cs.LG 89%

An Empirical Study of Sample Selection Strategies for Large Language Model Repair

Xuran Li, Jingyi Wang

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)；LLM(abstract)；分类 cs.LG

2510.19506 2025-10-23 cs.CL 89%

Lookahead Routing for Large Language Models

Canbin Huang, Tianyuan Shi, Yuhua Zhu, Ruijun Chen, Xiaojun Quan

机构 * School of Computer Science and Engineering, Sun Yat-sen University, China（中山大学计算机科学与工程学院）； Shenzhen Loop Area Institute, China（深圳环湖研究所）

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

2510.19389 2025-10-23 cs.LG 89%

ARA: Adaptive Rank Allocation for Efficient Large Language Model SVD Compression

Lin Xv, Jingsheng Gao, Xian Gao, Ting Liu, Yuzhuo Fu

机构 * Shanghai Jiao Tong University（上海交通大学）

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

2404.12494 2025-10-22 cs.CL 89%

BIRD: A Trustworthy Bayesian Inference Framework for Large Language Models

Yu Feng, Ben Zhou, Weidong Lin, Dan Roth

机构 * University of Pennsylvania（宾夕法尼亚大学）； Arizona State University（亚利桑那州立大学）

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

Journal ref ICLR 2025 (Oral)

2510.15304 2025-10-20 cs.CV cs.LG 89%

Layer as Puzzle Pieces: Compressing Large Language Models through Layer Concatenation

Fei Wang, Li Shen, Liang Ding, Chao Xue, Ye Liu, Changxing Ding

机构 * South China University of Technology（华南理工大学）； JD Explore Academy（JD探索学院）； Shenzhen Campus of Sun Yat-sen University（中山大学深圳校区）； University of Sydney（悉尼大学）； Pazhou Lab（琶洲实验室）

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

2510.14557 2025-10-17 cs.LG cs.AR 89%

MX+: Pushing the Limits of Microscaling Formats for Efficient Large Language Model Serving

Jungi Lee, Junyong Park, Soohyun Cha, Jaehoon Cho, Jaewoong Sim

机构 * Seoul National University（首尔国立大学）

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

Comments To appear at the 58th International Symposium on Microarchitecture (MICRO 2025)

2506.10597 2025-10-17 cs.CR cs.AI 89%

SoK: Evaluating Jailbreak Guardrails for Large Language Models

Xunguang Wang, Zhenlan Ji, Wenxuan Wang, Zongjie Li, Daoyuan Wu, Shuai Wang

机构 * The Hong Kong University of Science and Technology（香港科学与技术大学）； Renmin University of China（中国人民大学）； Lingnan University（岭南大学）

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

Comments Accepted by IEEE S&P 2026 Cycle 1

2510.12015 2025-10-15 cs.AI 89%

Asking Clarifying Questions for Preference Elicitation With Large Language Models

Ali Montazeralghaem, Guy Tennenholtz, Craig Boutilier, Ofer Meshi

机构 * Google（谷歌）

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

2501.15081 2025-10-10 cs.NE cs.AI 89%

Can Large Language Models Be Trusted as Evolutionary Optimizers for Network-Structured Combinatorial Problems?

Jie Zhao, Tao Wen, Kang Hao Cheong

机构 * Division of Mathematical Sciences, School of Physical and Mathematical Sciences, Nanyang Technological University（数学科学系，物理与数学科学学院，南洋理工大学）； School of Computer Science and Engineering, Nanyang Technological University（计算机科学与工程学院，南洋理工大学）

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

2510.05542 2025-10-08 cs.SD cs.CL eess.AS 89%

Sci-Phi: A Large Language Model Spatial Audio Descriptor

Xilin Jiang, Hannes Gamper, Sebastian Braun

机构 * Columbia University（哥伦比亚大学）； Microsoft Research（微软研究院）

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

2510.02752 2025-10-06 cs.CL 89%

The Path of Self-Evolving Large Language Models: Achieving Data-Efficient Learning via Intrinsic Feedback

Hangfan Zhang, Siyuan Xu, Zhimeng Guo, Huaisheng Zhu, Shicheng Liu, Xinrun Wang, Qiaosheng Zhang, Yang Chen, Peng Ye, Lei Bai, Shuyue Hu

机构 * Pennsylvania State University（宾夕法尼亚州立大学）； Singapore Management University（新加坡管理大学）； Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）

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

2510.02669 2025-10-06 cs.AI cs.HC cs.IR 89%

AutoMaAS: Self-Evolving Multi-Agent Architecture Search for Large Language Models

Bo Ma, Hang Li, ZeHua Hu, XiaoFan Gui, LuYao Liu, Simon Liu

机构 * Department of Software \& Microelectronics Peking University Beijing, China ； Department of Software \& Microelectronics Peking University Beijing, China hangli\ ； Department of Software \& Microelectronics Peking University Beijing, China zehua\ ； Department of Software \& Microelectronics Peking University Beijing, China xiaofan\ ； Economic Law School China University of Political Science ； School of Computer Science Peking University Beijing, China

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

1. 效率与部署 22405 篇

Collaborative Large Language Model Inference via Resource-Aware Parallel Speculative Decoding

A Survey on Inference Engines for Large Language Models: Perspectives on Optimization and Efficiency

Position-Aware Depth Decay Decoding ($D^3$): Boosting Large Language Model Inference Efficiency

E$^3$-Pruner: Towards Efficient, Economical, and Effective Layer Pruning for Large Language Models

Detecting Sleeper Agents in Large Language Models via Semantic Drift Analysis

AutoTool: Efficient Tool Selection for Large Language Model Agents

Simultaneous Machine Translation with Large Language Models

Leveraging Large Language Models for Use Case Model Generation from Software Requirements

EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language Models

ConfGuard: A Simple and Effective Backdoor Detection for Large Language Models

Quality Over Quantity? LLM-Based Curation for a Data-Efficient Audio-Video Foundation Model

KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse

GraphChain: Large Language Models for Large-scale Graph Analysis via Tool Chaining

Utilizing Multilingual Encoders to Improve Large Language Models for Low-Resource Languages

Evaluating The Performance of Using Large Language Models to Automate Summarization of CT Simulation Orders in Radiation Oncology

Using Multi-modal Large Language Model to Boost Fireworks Algorithm's Ability in Settling Challenging Optimization Tasks

Inference-Time Reward Hacking in Large Language Models

1+1>2: A Synergistic Sparse and Low-Rank Compression Method for Large Language Models

An Empirical Study of Sample Selection Strategies for Large Language Model Repair

Lookahead Routing for Large Language Models

ARA: Adaptive Rank Allocation for Efficient Large Language Model SVD Compression

BIRD: A Trustworthy Bayesian Inference Framework for Large Language Models

Layer as Puzzle Pieces: Compressing Large Language Models through Layer Concatenation

MX+: Pushing the Limits of Microscaling Formats for Efficient Large Language Model Serving

SoK: Evaluating Jailbreak Guardrails for Large Language Models

Asking Clarifying Questions for Preference Elicitation With Large Language Models

Can Large Language Models Be Trusted as Evolutionary Optimizers for Network-Structured Combinatorial Problems?

Sci-Phi: A Large Language Model Spatial Audio Descriptor

The Path of Self-Evolving Large Language Models: Achieving Data-Efficient Learning via Intrinsic Feedback

AutoMaAS: Self-Evolving Multi-Agent Architecture Search for Large Language Models