语言大模型 / LLM - arXivDaily 专题

2505.17716 2025-05-26 cs.LG cs.MA 83%

Get Experience from Practice: LLM Agents with Record & Replay

Erhu Feng, Wenbo Zhou, Zibin Liu, Le Chen, Yunpeng Dong, Cheng Zhang, Yisheng Zhao, Dong Du, Zhichao Hua, Yubin Xia, Haibo Chen

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

2505.17071 2025-05-26 cs.CL 83%

What's in a prompt? Language models encode literary style in prompt embeddings

Raphaël Sarfati, Haley Moller, Toni J. B. Liu, Nicolas Boullé, Christopher Earls

机构 * Cornell University（康奈尔大学）； Yale University（耶鲁大学）； Imperial College London（伦敦帝国理工学院）

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

2505.14309 2025-05-21 cs.CL 83%

Studying the Role of Input-Neighbor Overlap in Retrieval-Augmented Language Models Training Efficiency

Ehsan Doostmohammadi, Marco Kuhlmann

机构 * Linköping University（林雪平大学）

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

2505.11221 2025-05-19 cs.LG 83%

Sample Efficient Reinforcement Learning via Large Vision Language Model Distillation

Donghoon Lee, Tung M. Luu, Younghwan Lee, Chang D. Yoo

机构 * Robotics Program KAIST（韩国釜山科学技术院机器人计划）； Electrical Engineering KAIST（韩国釜山科学技术院电子工程）

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

Comments 5 pages, ICASSP 2025. The first two authors are equally contributed

Journal ref ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

2410.14609 2025-05-19 cs.IR cs.CL 83%

DiSCo: LLM Knowledge Distillation for Efficient Sparse Retrieval in Conversational Search

Simon Lupart, Mohammad Aliannejadi, Evangelos Kanoulas

机构 * University of Amsterdam（阿姆斯特丹大学）

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

Comments 11 pages, 6 figures. SIGIR '25 Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval July 13--18, 2025 Padua, Italy

2505.09810 2025-05-16 cs.LG 83%

Lossless Compression for LLM Tensor Incremental Snapshots

Daniel Waddington, Cornel Constantinescu

机构 * IBM Research, Almaden Research Center（IBM研究院，阿姆德恩研究中心）

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

2505.08542 2025-05-14 cs.AI 83%

Guiding LLM-based Smart Contract Generation with Finite State Machine

Hao Luo, Yuhao Lin, Xiao Yan, Xintong Hu, Yuxiang Wang, Qiming Zeng, Hao Wang, Jiawei Jiang

机构 * School of Computer Science, Wuhan University（武汉大学计算机学院）； School of Cyber Science and Engineering, Wuhan University（武汉大学网络科学与工程学院）

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

2503.12533 2025-05-13 cs.RO cs.LG 83%

Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills

Haoqi Yuan, Yu Bai, Yuhui Fu, Bohan Zhou, Yicheng Feng, Xinrun Xu, Yi Zhan, Börje F. Karlsson, Zongqing Lu

机构 * Peking University（北京大学）； BAAI（百度人工智能研究院）； BeingBeyond

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

2505.02138 2025-05-07 cs.LG 83%

Efficient Multivariate Time Series Forecasting via Calibrated Language Models with Privileged Knowledge Distillation

Chenxi Liu, Hao Miao, Qianxiong Xu, Shaowen Zhou, Cheng Long, Yan Zhao, Ziyue Li, Rui Zhao

机构 * S-Lab, Nanyang Technological University, Singapore（南洋理工大学S实验室）； Aalborg University, Denmark（奥尔堡大学）； University of Cologne, Germany（科隆大学）； SenseTime Research, China（SenseTime研究院）

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

Comments Accepted by ICDE 2025

2505.01050 2025-05-05 cs.CV cs.LG 83%

Transferable Adversarial Attacks on Black-Box Vision-Language Models

Kai Hu, Weichen Yu, Li Zhang, Alexander Robey, Andy Zou, Chengming Xu, Haoqi Hu, Matt Fredrikson

机构 * Carnegie Mellon University（卡内基梅隆大学）

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

2501.18265 2025-05-02 cs.IR cs.CL cs.HC 83%

Efficiency and Effectiveness of LLM-Based Summarization of Evidence in Crowdsourced Fact-Checking

Kevin Roitero, Dustin Wright, Michael Soprano, Isabelle Augenstein, Stefano Mizzaro

机构 * Department of Computer Science, University of Udine（乌迪大学计算机科学系）； Department of Computer Science, University of Copenhagen（哥本哈根大学计算机科学系）

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

Comments 19 pages; 7 figures; 5 tables

2504.21553 2025-05-01 cs.CL 83%

Precision Where It Matters: A Novel Spike Aware Mixed-Precision Quantization Strategy for LLaMA-based Language Models

Lucas Maisonnave, Cyril Moineau, Olivier Bichler, Fabrice Rastello

机构 * Univ. Paris-Saclay, CEA LIST（巴黎萨克雷大学，CEA LIST）； Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP, LIG（格勒诺布尔阿尔卑斯大学，法国国家科学研究中心，格勒诺布尔INP，LIG）

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

2504.19449 2025-04-29 cs.LG 83%

R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference

Zhenyu Zhang, Zechun Liu, Yuandong Tian, Harshit Khaitan, Zhangyang Wang, Steven Li

机构 * The University of Texas at Austin（德克萨斯大学奥斯汀分校）； Meta AI

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

Comments ICLR 2025

2405.19325 2025-04-28 cs.CL 83%

Nearest Neighbor Speculative Decoding for LLM Generation and Attribution

Minghan Li, Xilun Chen, Ari Holtzman, Beidi Chen, Jimmy Lin, Wen-tau Yih, Xi Victoria Lin

机构 * Cohere ； Meta FAIR ； University of Chicago（芝加哥大学）； Carnegie Mellon University（卡内基梅隆大学）； University of Waterloo（滑铁卢大学）

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

Journal ref Advances in Neural Information Processing Systems (2024), vol. 37, page 80987-81015

2504.17376 2025-04-25 cs.AR cs.LG 83%

On-Device Qwen2.5: Efficient LLM Inference with Model Compression and Hardware Acceleration

Maoyang Xiang, Ramesh Fernando, Bo Wang

机构 * Singapore University of Technology and Design（新加坡科技设计大学）

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

2504.15299 2025-04-23 cs.DC cs.AI 83%

D$^{2}$MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving

Haodong Wang, Qihua Zhou, Zicong Hong, Song Guo

机构 * Hong Kong University of Science and Technology（香港理工大学）； College of Computer Science and Software Engineering, Shenzhen University（深圳大学计算机科学与软件工程学院）

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

Comments Accepted by MobiCom 2025

2504.15027 2025-04-22 cs.CL 83%

DistilQwen2.5: Industrial Practices of Training Distilled Open Lightweight Language Models

Chengyu Wang, Junbing Yan, Yuanhao Yue, Jun Huang

机构 * Alibaba Cloud Computing（阿里巴巴云 computing）

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

2504.09345 2025-04-15 cs.DC cs.AI 83%

MoE-Lens: Towards the Hardware Limit of High-Throughput MoE LLM Serving Under Resource Constraints

Yichao Yuan, Lin Ma, Nishil Talati

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

2409.01990 2025-04-15 cs.DC cs.LG 83%

Designing Large Foundation Models for Efficient Training and Inference: A Survey

Dong Liu, Yanxuan Yu, Yite Wang, Jing Wu, Zhongwei Wan, Sina Alinejad, Benjamin Lengerich, Ying Nian Wu

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

2504.08165 2025-04-14 cs.CL 83%

Findings of the BabyLM Challenge: Sample-Efficient Pretraining on Developmentally Plausible Corpora

Alex Warstadt, Aaron Mueller, Leshem Choshen, Ethan Wilcox, Chengxu Zhuang, Juan Ciro, Rafael Mosquera, Bhargavi Paranjape, Adina Williams, Tal Linzen, Ryan Cotterell

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

Comments Published in Proceedings of BabyLM. Please cite the published version on ACL anthology: http://aclanthology.org/2023.conll-babylm.1/

Journal ref 2023. In Proceedings of the BabyLM Challenge at the 27th Conference on Computational Natural Language Learning, pages 1-34, Singapore. Association for Computational Linguistics

2504.04698 2025-04-08 cs.CL 83%

scAgent: Universal Single-Cell Annotation via a LLM Agent

Yuren Mao, Yu Mi, Peigen Liu, Mengfei Zhang, Hanqing Liu, Yunjun Gao

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

2407.20143 2025-04-03 cs.AI 83%

ByteCheckpoint: A Unified Checkpointing System for Large Foundation Model Development

Borui Wan, Mingji Han, Yiyao Sheng, Yanghua Peng, Haibin Lin, Mofan Zhang, Zhichao Lai, Menghan Yu, Junda Zhang, Zuquan Song, Xin Liu, Chuan Wu

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

2503.23294 2025-04-01 cs.CL 83%

Cocktail: Chunk-Adaptive Mixed-Precision Quantization for Long-Context LLM Inference

Wei Tao, Bin Zhang, Xiaoyang Qu, Jiguang Wan, Jianzong Wang

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

Comments Accepted by the Design, Automation, and Test in Europe 2025 (DATE 2025)

2502.09054 2025-04-01 cs.AI 83%

Cost-Saving LLM Cascades with Early Abstention

Michael J. Zellinger, Rex Liu, Matt Thomson

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

Comments 9 pages, 3 figures

2411.09587 2025-03-20 cs.CL 83%

BabyLM Challenge: Exploring the Effect of Variation Sets on Language Model Training Efficiency

Akari Haga, Akiyo Fukatsu, Miyu Oba, Arianna Bisazza, Yohei Oseki

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

Comments Accepted by BabyLM challenge 2024 at CONLL 2024 ( https://aclanthology.org/2024.conll-babylm.23 )

2409.16560 2025-03-17 cs.AI 83%

Dynamic-Width Speculative Beam Decoding for Efficient LLM Inference

Zongyue Qin, Zifan He, Neha Prakriya, Jason Cong, Yizhou Sun

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

2410.18403 2025-03-14 q-bio.BM cs.LG 83%

Structure Language Models for Protein Conformation Generation

Jiarui Lu, Xiaoyin Chen, Stephen Zhewen Lu, Chence Shi, Hongyu Guo, Yoshua Bengio, Jian Tang

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

Comments Published as a conference paper at ICLR 2025, see https://openreview.net/forum?id=OzUNDnpQyd

2411.15232 2025-03-13 cs.CV cs.CL 83%

BiomedCoOp: Learning to Prompt for Biomedical Vision-Language Models

Taha Koleilat, Hojat Asgariandehkordi, Hassan Rivaz, Yiming Xiao

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

Comments Accepted to CVPR 2025

2503.07993 2025-03-12 cs.AI 83%

LLM-Powered Knowledge Graphs for Enterprise Intelligence and Analytics

Rajeev Kumar, Kumar Ishan, Harishankar Kumar, Abhinandan Singla

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

2503.04675 2025-03-07 cs.CL 83%

LLM-guided Plan and Retrieval: A Strategic Alignment for Interpretable User Satisfaction Estimation in Dialogue

Sangyeop Kim, Sohhyung Park, Jaewon Jung, Jinseok Kim, Sungzoon Cho

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

Comments Accepted by NAACL 2025

1. 效率与部署 22456 篇

Get Experience from Practice: LLM Agents with Record & Replay

What's in a prompt? Language models encode literary style in prompt embeddings

Studying the Role of Input-Neighbor Overlap in Retrieval-Augmented Language Models Training Efficiency

Sample Efficient Reinforcement Learning via Large Vision Language Model Distillation

DiSCo: LLM Knowledge Distillation for Efficient Sparse Retrieval in Conversational Search

Lossless Compression for LLM Tensor Incremental Snapshots

Guiding LLM-based Smart Contract Generation with Finite State Machine

Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills

Efficient Multivariate Time Series Forecasting via Calibrated Language Models with Privileged Knowledge Distillation

Transferable Adversarial Attacks on Black-Box Vision-Language Models

Efficiency and Effectiveness of LLM-Based Summarization of Evidence in Crowdsourced Fact-Checking

Precision Where It Matters: A Novel Spike Aware Mixed-Precision Quantization Strategy for LLaMA-based Language Models

R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference

Nearest Neighbor Speculative Decoding for LLM Generation and Attribution

On-Device Qwen2.5: Efficient LLM Inference with Model Compression and Hardware Acceleration

D$^{2}$MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving

DistilQwen2.5: Industrial Practices of Training Distilled Open Lightweight Language Models

MoE-Lens: Towards the Hardware Limit of High-Throughput MoE LLM Serving Under Resource Constraints

Designing Large Foundation Models for Efficient Training and Inference: A Survey

Findings of the BabyLM Challenge: Sample-Efficient Pretraining on Developmentally Plausible Corpora

scAgent: Universal Single-Cell Annotation via a LLM Agent

ByteCheckpoint: A Unified Checkpointing System for Large Foundation Model Development

Cocktail: Chunk-Adaptive Mixed-Precision Quantization for Long-Context LLM Inference

Cost-Saving LLM Cascades with Early Abstention

BabyLM Challenge: Exploring the Effect of Variation Sets on Language Model Training Efficiency

Dynamic-Width Speculative Beam Decoding for Efficient LLM Inference

Structure Language Models for Protein Conformation Generation

BiomedCoOp: Learning to Prompt for Biomedical Vision-Language Models

LLM-Powered Knowledge Graphs for Enterprise Intelligence and Analytics

LLM-guided Plan and Retrieval: A Strategic Alignment for Interpretable User Satisfaction Estimation in Dialogue