语言大模型 / LLM - arXivDaily 专题

2509.04377 2025-09-05 cs.LG 89%

PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference

Krishna Teja Chitty-Venkata, Jie Ye, Xian-He Sun, Anthony Kougkas, Murali Emani, Venkatram Vishwanath, Bogdan Nicolae

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 Preprint

2501.12689 2025-09-05 cs.LG 89%

IC-Cache: Efficient Large Language Model Serving via In-context Caching

Yifan Yu, Yu Gan, Nikhil Sarda, Lillian Tsai, Jiaming Shen, Yanqi Zhou, Arvind Krishnamurthy, Fan Lai, Henry M. Levy, David Culler

机构 * University of Illinois Urbana-Champaign ； Google ； Google \& University of Washington

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

2509.03837 2025-09-05 cs.LG cs.IT math.IT 88%

Vehicle-to-Infrastructure Collaborative Spatial Perception via Multimodal Large Language Models

Kimia Ehsani, Walid Saad

机构 * Bradley Department of Electrical and Computer Engineering（电气与计算机工程系）

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

Comments Accepted at IEEE GLOBECOM 2025

2509.03533 2025-09-05 cs.CL cs.LG q-fin.GN 86%

Topic Identification in LLM Input-Output Pairs through the Lens of Information Bottleneck

Igor Halperin

机构 * Fidelity Investments（fidelity投资公司）

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

Comments 26 pages, 4 figures

2509.03890 2025-09-05 cs.AI 85%

FaMA: LLM-Empowered Agentic Assistant for Consumer-to-Consumer Marketplace

Yineng Yan, Xidong Wang, Jin Seng Cheng, Ran Hu, Wentao Guan, Nahid Farahmand, Hengte Lin, Yue Li

机构 * University of Texas at Austin（德克萨斯大学奥斯汀分校）； Meta Platforms, Inc（Meta平台公司）

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

2410.01228 2025-09-05 cs.DC cs.LG 85%

ConServe: Fine-Grained GPU Harvesting for LLM Online and Offline Co-Serving

Yifan Qiao, Shu Anzai, Shan Yu, Haoran Ma, Shuo Yang, Yang Wang, Miryung Kim, Yongji Wu, Yang Zhou, Jiarong Xing, Joseph E. Gonzalez, Ion Stoica, Harry Xu

机构 * UC Berkeley（伯克利大学）； UCLA（洛杉矶大学）； Rice University（里奇大学）； UC Davis（戴维斯大学）； Intel（英特尔公司）

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

2506.07900 2025-09-05 cs.CL cs.AI 82%

MiniCPM4: Ultra-Efficient LLMs on End Devices

MiniCPM Team, Chaojun Xiao, Yuxuan Li, Xu Han, Yuzhuo Bai, Jie Cai, Haotian Chen, Wentong Chen, Xin Cong, Ganqu Cui, Ning Ding, Shengda Fan, Yewei Fang, Zixuan Fu, Wenyu Guan, Yitong Guan, Junshao Guo, Yufeng Han, Bingxiang He, Yuxiang Huang, Baoxi Ji, Cunliang Kong, Qiuzuo Li, Siyuan Li, Wenhao Li, Xin Li, Yanghao Li, Yishan Li, Zhen Li, Dan Liu, Biyuan Lin, Yankai Lin, Xiang Long, Quanyu Lu, Yaxi Lu, Peiyan Luo, Hongya Lyu, Litu Ou, Yinxu Pan, Lushi Pu, Zekai Qu, Qundong Shi, Zijun Song, Jiayuan Su, Zhou Su, Ao Sun, Xianghui Sun, Peijun Tang, Fangzheng Wang, Feng Wang, Shuo Wang, Yudong Wang, Zheng Wang, Yesai Wu, Zhenyu Xiao, Jie Xie, Zihao Xie, Xiaoyue Xu, Yukun Yan, Jiarui Yuan, Jinqian Zhang, Kaihuo Zhang, Lei Zhang, Linyue Zhang, Xueren Zhang, Yudi Zhang, Hengyu Zhao, Weilin Zhao, Weilun Zhao, Yuanqian Zhao, Zhi Zheng, Chuyue Zhou, Ge Zhou, Jie Zhou, Wei Zhou, Yanghao Zhou, Zihan Zhou, Zixuan Zhou, Zhiyuan Liu, Guoyang Zeng, Chao Jia, Dahai Li, Maosong Sun

机构 * MiniCPM Team（MiniCPM团队）

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

Comments MiniCPM4 Technical Report

2502.14791 2025-09-05 cs.CL cs.AI cs.LG 80%

Rapid Word Learning Through Meta In-Context Learning

Wentao Wang, Guangyuan Jiang, Tal Linzen, Brenden M. Lake

机构 * New York University（纽约大学）； MIT（麻省理工学院）

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

Comments EMNLP 2025

2509.04185 2025-09-05 cs.LG 79%

Set Block Decoding is a Language Model Inference Accelerator

Itai Gat, Heli Ben-Hamu, Marton Havasi, Daniel Haziza, Jeremy Reizenstein, Gabriel Synnaeve, David Lopez-Paz, Brian Karrer, Yaron Lipman

机构 * FAIR at Meta（Meta的FAIR）

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

2509.03746 2025-09-05 cs.IR 78%

Efficient Item ID Generation for Large-Scale LLM-based Recommendation

Anushya Subbiah, Vikram Aggarwal, James Pine, Steffen Rendle, Krishna Sayana, Kun Su

专题命中效率与部署：LLM(title,abstract)

2509.03956 2025-09-05 cs.AI 77%

World Model Implanting for Test-time Adaptation of Embodied Agents

Minjong Yoo, Jinwoo Jang, Sihyung Yoon, Honguk Woo

机构 * Department of Computer Science and Engineering, Sungkyunkwan University（计算机科学与工程系，顺天大学）

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

Journal ref Forty-second International Conference on Machine Learning, 2025

2509.00073 2025-09-05 cs.LG 77%

Mitigating Clinician Information Overload: Generative AI for Integrated EHR and RPM Data Analysis

Ankit Shetgaonkar, Dipen Pradhan, Lakshit Arora, Sanjay Surendranath Girija, Shashank Kapoor, Aman Raj

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

Comments Accepted at IEEE COMPSAC 2025

Journal ref 2025 IEEE 49th Annual Computers, Software, and Applications Conference (COMPSAC)

2509.04356 2025-09-05 cs.HC cs.RO 75%

SRWToolkit: An Open Source Wizard of Oz Toolkit to Create Social Robotic Avatars

Atikkhan Faridkhan Nilgar, Kristof Van Laerhoven, Ayub Kinoti

机构 * University of Siegen（施皮格恩大学）； Honda Research Institute Europe GmbH（本田欧洲研究院）； Dedan Kimathi University of Technology（德丹·基马蒂技术大学）

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

Journal ref 2025 International Conference on Social Robotics (ICSR)

2505.05122 2025-09-05 cs.DB cs.LG 70%

Text2Cypher: Data Pruning using Hard Example Selection

Makbule Gulcin Ozsoy

机构 * Neo4j

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

2505.05118 2025-09-05 cs.DB cs.LG 70%

Enhancing Text2Cypher with Schema Filtering

Makbule Gulcin Ozsoy

机构 * Neo4j

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

2509.04326 2025-09-05 cs.CV 67%

Efficient Odd-One-Out Anomaly Detection

Silvio Chito, Paolo Rabino, Tatiana Tommasi

机构 * Politecnico di Torino（托斯尼亚理工学院）

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

Comments Accepted at ICIAP 2025

2509.03615 2025-09-05 cs.CL cs.AI 62%

E-ARMOR: Edge case Assessment and Review of Multilingual Optical Character Recognition

Aryan Gupta, Anupam Purwar

机构 * Sprinklr

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

Comments Sprinklr OCR provides a fast and compute light way of performing OCR

2509.04180 2025-09-05 cs.CV cs.AI 57%

VisioFirm: Cross-Platform AI-assisted Annotation Tool for Computer Vision

Safouane El Ghazouali, Umberto Michelucci

机构 * TOELT LLC AI lab（TOELT LLC人工智能实验室）

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

2509.04162 2025-09-05 cs.AR 50%

Real Time FPGA Based Transformers & VLMs for Vision Tasks: SOTA Designs and Optimizations

Safa Mohammed Sali, Mahmoud Meribout, Ashiyana Abdul Majeed

专题命中效率与部署：language model(abstract)

1. 效率与部署 19 篇

PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference

IC-Cache: Efficient Large Language Model Serving via In-context Caching

Vehicle-to-Infrastructure Collaborative Spatial Perception via Multimodal Large Language Models

Topic Identification in LLM Input-Output Pairs through the Lens of Information Bottleneck

FaMA: LLM-Empowered Agentic Assistant for Consumer-to-Consumer Marketplace

ConServe: Fine-Grained GPU Harvesting for LLM Online and Offline Co-Serving

MiniCPM4: Ultra-Efficient LLMs on End Devices

Rapid Word Learning Through Meta In-Context Learning

Set Block Decoding is a Language Model Inference Accelerator

Efficient Item ID Generation for Large-Scale LLM-based Recommendation

World Model Implanting for Test-time Adaptation of Embodied Agents

Mitigating Clinician Information Overload: Generative AI for Integrated EHR and RPM Data Analysis

SRWToolkit: An Open Source Wizard of Oz Toolkit to Create Social Robotic Avatars

Text2Cypher: Data Pruning using Hard Example Selection

Enhancing Text2Cypher with Schema Filtering

Efficient Odd-One-Out Anomaly Detection

E-ARMOR: Edge case Assessment and Review of Multilingual Optical Character Recognition

VisioFirm: Cross-Platform AI-assisted Annotation Tool for Computer Vision

Real Time FPGA Based Transformers & VLMs for Vision Tasks: SOTA Designs and Optimizations