语言大模型 / LLM - arXivDaily 专题

2304.05501 2023-12-27 cs.RO 89%

L3MVN: Leveraging Large Language Models for Visual Target Navigation

Bangguo Yu, Hamidreza Kasaei, Ming Cao

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

Comments 7 pages

Journal ref 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

2312.12464 2023-12-22 cs.LG cs.AI cs.CL 89%

Towards Better Serialization of Tabular Data for Few-shot Classification with Large Language Models

Sukriti Jaitly, Tanay Shah, Ashish Shugani, Razik Singh Grewal

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

Comments 4 pages, 2 figures

2312.13225 2023-12-21 cs.SE 89%

Automated DevOps Pipeline Generation for Code Repositories using Large Language Models

Deep Mehta, Kartik Rawool, Subodh Gujar, Bowen Xu

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

2309.02784 2023-12-14 cs.LG cs.AI cs.CL 89%

Norm Tweaking: High-performance Low-bit Quantization of Large Language Models

Liang Li, Qingyuan Li, Bo Zhang, Xiangxiang Chu

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

2311.17474 2023-11-30 cs.NI 89%

Large Language Models for Networking: Applications, Enabling Techniques, and Challenges

Yudong Huang, Hongyang Du, Xinyuan Zhang, Dusit Niyato, Jiawen Kang, Zehui Xiong, Shuo Wang, Tao Huang

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

Comments 7 pages, 3 figures, 2 tables

2311.01266 2023-11-03 cs.SE 89%

Let's Discover More API Relations: A Large Language Model-based AI Chain for Unsupervised API Relation Inference

Qing Huang, Yanbang Sun, Zhenchang Xing, Yuanlong Cao, Jieshan Chen, Xiwei Xu, Huan Jin, Jiaxing Lu

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

2310.18356 2023-11-01 cs.CL cs.AI cs.LG 89%

LoRAShear: Efficient Large Language Model Structured Pruning and Knowledge Recovery

Tianyi Chen, Tianyu Ding, Badal Yadav, Ilya Zharkov, Luming Liang

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

2310.04698 2023-10-10 cs.CV 89%

Tree-GPT: Modular Large Language Model Expert System for Forest Remote Sensing Image Understanding and Interactive Analysis

Siqi Du, Shengjun Tang, Weixi Wang, Xiaoming Li, Renzhong Guo

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

2310.01612 2023-10-04 cs.IR 89%

Towards Efficient and Effective Adaptation of Large Language Models for Sequential Recommendation

Bo Peng, Ben Burns, Ziqi Chen, Srinivasan Parthasarathy, Xia Ning

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

2309.05248 2023-09-15 eess.AS cs.SD 89%

Enhancing Speaker Diarization with Large Language Models: A Contextual Beam Search Approach

Tae Jin Park, Kunal Dhawan, Nithin Koluguri, Jagadeesh Balam

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

Comments 4 pages 1 reference page, ICASSP format

2305.17256 2023-09-12 cs.CL cs.AI cs.LG 89%

Large Language Models Can be Lazy Learners: Analyze Shortcuts in In-Context Learning

Ruixiang Tang, Dehan Kong, Longtao Huang, Hui Xue

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

2308.11601 2023-08-24 cs.LG cs.AI cs.CL cs.MA 89%

Tryage: Real-time, intelligent Routing of User Prompts to Large Language Models

Surya Narayanan Hari, Matt Thomson

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

2308.10837 2023-08-22 cs.IR 89%

Leveraging Large Language Models for Pre-trained Recommender Systems

Zhixuan Chu, Hongyan Hao, Xin Ouyang, Simeng Wang, Yan Wang, Yue Shen, Jinjie Gu, Qing Cui, Longfei Li, Siqiao Xue, James Y Zhang, Sheng Li

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

Comments 13 pages, 4 figures

2303.04673 2023-08-10 cs.CL cs.AI cs.LG 89%

Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference

Chi Wang, Susan Xueqing Liu, Ahmed H. Awadallah

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

2305.15336 2023-05-25 cs.CR 89%

From Text to MITRE Techniques: Exploring the Malicious Use of Large Language Models for Generating Cyber Attack Payloads

P. V. Sai Charan, Hrushikesh Chunduri, P. Mohan Anand, Sandeep K Shukla

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

2304.07493 2023-04-18 cs.AR 89%

OliVe: Accelerating Large Language Models via Hardware-friendly Outlier-Victim Pair Quantization

Cong Guo, Jiaming Tang, Weiming Hu, Jingwen Leng, Chen Zhang, Fan Yang, Yunxin Liu, Minyi Guo, Yuhao Zhu

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

Comments ISCA 2023

2302.02094 2023-02-14 cs.HC 89%

Chat2VIS: Generating Data Visualisations via Natural Language using ChatGPT, Codex and GPT-3 Large Language Models

Paula Maddigan, Teo Susnjak

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

Comments revision

2204.06644 2022-04-19 cs.LG cs.AI cs.CL 89%

METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

Payal Bajaj, Chenyan Xiong, Guolin Ke, Xiaodong Liu, Di He, Saurabh Tiwary, Tie-Yan Liu, Paul Bennett, Xia Song, Jianfeng Gao

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

Comments Update details in scaled initialization and add acknowledgement

2606.21238 2026-06-23 cs.DC cs.AI cs.LG 新提交 89%

Recency/Frequency Adaptive KV Caching for Large Language Model Serving

面向大语言模型服务的近期/频率自适应KV缓存

Yang Shen, Meghana Madhyastha, Robert Underwood, Bogdan Nicolae, Randal Burns

机构 * Department of Computer Science, Johns Hopkins University, Baltimore, USA（约翰霍普金斯大学计算机科学系）； Argonne National Laboratory, Lemont, USA（阿贡国家实验室）； Parasail, Inc., San Mateo, USA（Parasail公司）

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

AI总结针对LRU策略在多工作负载下缓存失效问题，提出自适应KV缓存，动态分配近期与高频KV块的空间，在合成文档问答中缓存命中率提升10.8%，首令牌时间降低12.6%。

Comments Accepted at the ICML 2026 Workshop on Resource-Adaptive Foundation Model Inference (AdaptFM)

2604.24542 2026-04-28 cs.CR cs.AI cs.CL 89%

Layerwise Convergence Fingerprints for Runtime Misbehavior Detection in Large Language Models

用于大语言模型运行时恶意行为检测的分层收敛指纹

Nay Myat Min, Long H. Pham, Jun Sun

机构 * Singapore Management University（新加坡管理大学）

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

AI总结本文提出LCF方法，通过分析神经网络层间隐藏状态轨迹来检测运行时恶意行为，无需参考模型或重训练，有效降低攻击成功率并高检测率识别多种威胁。

Comments 34 pages, 5 figures. Code: https://github.com/NayMyatMin/LCF-LLM

2411.08610 2024-11-14 cs.CL cs.LG 89%

Dynamic Subset Tuning: Expanding the Operational Range of Parameter-Efficient Training for Large Language Models

Felix Stahlberg, Jared Lichtarge, Shankar Kumar

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

Comments NeurIPS 2024 Workshop on Adaptive Foundation Models

2312.07046 2023-12-13 cs.LG cs.CL 89%

Rethinking Compression: Reduced Order Modelling of Latent Features in Large Language Models

Arnav Chavan, Nahush Lele, Deepak Gupta

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

Comments Brief technical report; Code will be made available at https://github.com/transmuteAI/trailmet/tree/main/trailmet/algorithms/llm-rom

2608.19670 2026-08-21 cs.CL 新提交 89%

The Asymmetric Harms of LLM Compression

大语言模型压缩的非对称危害

Yuan Wu, Mairui Li, Lesia Semenova, Chudi Zhong

机构 * Rice University（莱斯大学）； University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）； Rutgers University（罗格斯大学）

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

AI总结本研究通过评估3个LLM、11种压缩方法，发现LLM压缩会引发知识保留、置信度及偏见的非对称变化，聚合指标无法捕捉，需对压缩模型开展细粒度评估。

2608.17336 2026-08-19 cs.AI 新提交 89%

TileMix: Tile-Centric Mixed-Precision Attention for LLM Inference Acceleration

TileMix：面向大语言模型推理加速的以块为中心的混合精度注意力机制

Hanzhi Zhang, Qiao Zhang, Qinglei Cao, Heng Fan, Yan Huang, Kewei Sha, Yunhe Feng

机构 * University of North Texas（北得克萨斯大学）； Saint Louis University（圣路易斯大学）

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

AI总结 TileMix是一种以块为中心的混合精度注意力内核，通过硬件对齐的分数块路由实现LLM推理加速，可恢复长上下文质量并提升预填充吞吐量，支持多种模型与任务场景。

2608.08156 2026-08-11 cs.LG cs.NE 新提交 89%

A Hybrid Nested Harness for Decoupling Structure and Parameters in LLM-Driven Optimization

用于解耦大语言模型驱动优化中结构与参数的混合嵌套框架

Víctor Gallego

机构 * Komorebi AI Technologies（Komorebi人工智能技术公司）

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

AI总结该研究提出混合嵌套框架解耦LLM驱动优化的结构与参数，外层LLM提结构草图、内层优化器调参，在三类任务中性能优于纯LLM及纯数值优化基线。

Comments Published as a conference paper at LM4Sci Workshop @ COLM 2026

2608.07427 2026-08-10 cs.AI cs.PF 新提交 89%

A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy

一图胜千词：视觉语言模型如何在提升准确率的同时降低AI能源成本

Bhavika Jalli, Nikhil Korati Prasanna, Jayanta Choudhury

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

AI总结该研究提出将时间序列编码为二维图的视觉语言模型，可大幅减少输入词元、降低推理能耗，同时在电信异常检测等任务中提升准确率，解决了LLM处理多维度KPI数据的低效问题。

Comments Accepted at the 14th European Conference on Renewable Energy Systems (ECRES), July 7--9, 2026, London, UK

2509.10317 2026-08-10 cs.RO cs.LG 版本更新 89%

Robot guide with multi-agent control and automatic scenario generation with LLM

基于多智能体控制与大语言模型自动场景生成的机器人引导系统

Elizaveta D. Moskovskaya, Anton D. Moscowsky

机构 * National Research Center "Kurchatov Institute"（国家研究快机构"库恰托夫研究所"）

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

AI总结该研究开发了结合多智能体资源管理与LLM自动场景生成的混合社交机器人控制架构，在MENTOR-1导游机器人上验证了其能提升长期讲故事任务中机器人交互的自然性与丰富度。

Comments 14 pages, 4 figures, 4 tables, 1 demo-video and repository link. There were major changes: an introduction, a review, and a new experiment. Some tables and figures have also been changed

2608.06135 2026-08-07 cs.LG 新提交 89%

LLM Inference Under Bursty Workload Distribution: Modifying the WAIT Algorithm

突发工作负载分布下的大语言模型推理：对WAIT算法的改进

Anjali Gangadhar Katageria, Shobha Rani, Raghu Nandan Sengupta

机构 * Indian Institute of Technology Dharwad（印度达尔瓦德印度理工学院）； Indian Institute of Technology Kanpur（印度坎普尔印度理工学院）

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

AI总结该研究针对LLM推理中突发工作负载的问题，改进了WAIT算法，通过在线估计请求强度提升了低到达率变化场景下的吞吐量，且延迟与现有基准相当。

2602.11083 2026-08-04 cs.LG cs.CR 版本更新 89%

Token-Efficient Change Detection in LLM APIs

LLM API中的令牌高效变化检测

Timothée Chauvin, Clément Lalanne, Erwan Le Merrer, Jean-Michel Loubes, François Taïani, Gilles Tredan

机构 * Université de Rennes, Inria, CNRS/IRISA（里昂大学、法国国家信息与自动化技术研究院、法国国家科学研究中心/IRISA）； Equipe Regalia, Inria（Regalia团队、法国国家信息与自动化技术研究院）； LAAS, CNRS（拉劳斯研究中心、法国国家科学研究中心）； Univ Toulouse, INUC, UT2J, INSA Toulouse, TSE, CNRS, IMT（图卢兹大学、INUC、UT2J、图卢兹国家高等工业学院、TSE、法国国家科学研究中心、IMT）

专题命中效率与部署：LLM(title,title_cn)；分类 cs.LG

AI总结提出基于边界输入的黑盒变化检测方案B3IT，在仅观察输出令牌的条件下实现低成本、高性能的LLM变化检测。

Comments ICML 2026

2607.28966 2026-08-03 cs.CL 新提交 89%

BLADE: Boundary-Expanded and Layer-Adaptive Dynamic Exit for Efficient LLM Reasoning

BLADE：用于高效大语言模型推理的边界扩展与层自适应动态出口

Keshu Fu, Keqin Peng, Jun Bai, Shuhan Qin, Chen Li, Junzhu Liang, Yefei Chen, Jiaqi Li, Yuanxin Ouyang

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

AI总结 BLADE是一种轻量级框架，通过构建多粒度检查点、学习紧凑探测层子集等，在保持接近基线准确率的同时，减少大语言模型的推理计算与生成令牌量，提升了LLM推理效率。

Comments 8 pages

1. 效率与部署 22405 篇

L3MVN: Leveraging Large Language Models for Visual Target Navigation

Towards Better Serialization of Tabular Data for Few-shot Classification with Large Language Models

Automated DevOps Pipeline Generation for Code Repositories using Large Language Models

Norm Tweaking: High-performance Low-bit Quantization of Large Language Models

Large Language Models for Networking: Applications, Enabling Techniques, and Challenges

Let's Discover More API Relations: A Large Language Model-based AI Chain for Unsupervised API Relation Inference

LoRAShear: Efficient Large Language Model Structured Pruning and Knowledge Recovery

Tree-GPT: Modular Large Language Model Expert System for Forest Remote Sensing Image Understanding and Interactive Analysis

Towards Efficient and Effective Adaptation of Large Language Models for Sequential Recommendation

Enhancing Speaker Diarization with Large Language Models: A Contextual Beam Search Approach

Large Language Models Can be Lazy Learners: Analyze Shortcuts in In-Context Learning

Tryage: Real-time, intelligent Routing of User Prompts to Large Language Models

Leveraging Large Language Models for Pre-trained Recommender Systems

Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference

From Text to MITRE Techniques: Exploring the Malicious Use of Large Language Models for Generating Cyber Attack Payloads

OliVe: Accelerating Large Language Models via Hardware-friendly Outlier-Victim Pair Quantization

Chat2VIS: Generating Data Visualisations via Natural Language using ChatGPT, Codex and GPT-3 Large Language Models

METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

Recency/Frequency Adaptive KV Caching for Large Language Model Serving

Layerwise Convergence Fingerprints for Runtime Misbehavior Detection in Large Language Models

Dynamic Subset Tuning: Expanding the Operational Range of Parameter-Efficient Training for Large Language Models

Rethinking Compression: Reduced Order Modelling of Latent Features in Large Language Models

The Asymmetric Harms of LLM Compression

TileMix: Tile-Centric Mixed-Precision Attention for LLM Inference Acceleration

A Hybrid Nested Harness for Decoupling Structure and Parameters in LLM-Driven Optimization

A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy

Robot guide with multi-agent control and automatic scenario generation with LLM

LLM Inference Under Bursty Workload Distribution: Modifying the WAIT Algorithm

Token-Efficient Change Detection in LLM APIs

BLADE: Boundary-Expanded and Layer-Adaptive Dynamic Exit for Efficient LLM Reasoning