语言大模型 / LLM - arXivDaily 专题

2509.06221 2025-09-09 eess.AS cs.AI cs.CL cs.HC 86%

Beamforming-LLM: What, Where and When Did I Miss?

Vishal Choudhari

机构 * Columbia University（哥伦比亚大学）； Institute for Clarity in Documentation（清晰文档研究所）； Inria Paris-Rocquencourt（Inria巴黎- Rocquencourt研究所）； Rajiv Gandhi University（拉贾·甘地大学）； Tsinghua University（清华大学）； Palmer Research Laboratories（帕勒姆研究实验室）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2406.17642 2025-09-04 cs.CL cs.AI 86%

Banishing LLM Hallucinations Requires Rethinking Generalization

Johnny Li, Saksham Consul, Eda Zhou, James Wong, Naila Farooqui, Yuxin Ye, Nithyashree Manohar, Zhuxiaona Wei, Tian Wu, Ben Echols, Sharon Zhou, Gregory Diamos

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments I want to revisit some of the experiments in this paper, specifically figure 5

2509.00510 2025-09-03 cs.AI cs.CL 86%

LLM-Assisted Iterative Evolution with Swarm Intelligence Toward SuperBrain

Li Weigang, Pedro Carvalho Brom, Lucas Ramson Siefert

机构 * TransLab, Computer Science Department University of Brasilia（TransLab，计算机科学系巴西大学）； Math Department Federal Institute of Brasilia（数学系巴西联邦理工学院）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 24 pages, 5 figures

2502.21231 2025-09-01 cs.DC cs.AI cs.LG 86%

ByteScale: Efficient Scaling of LLM Training with a 2048K Context Length on More Than 12,000 GPUs

Hao Ge, Junda Feng, Qi Huang, Fangcheng Fu, Xiaonan Nie, Lei Zuo, Haibin Lin, Bin Cui, Xin Liu

机构 * Peking University（北京大学）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments 12 pages, 21 figures

Journal ref SIGCOMM 2025

2507.11742 2025-08-26 cs.CL cs.AI 86%

CRABS: A syntactic-semantic pincer strategy for bounding LLM interpretation of Python notebooks

Meng Li, Timothy M. McPhillips, Dingmin Wang, Shin-Rong Tsai, Bertram Ludäscher

机构 * School of Information Sciences, University of Illinois Urbana-Champaign（信息科学学院，伊利诺伊大学厄巴纳-香槟分校）； Department of Computer Science, University of Oxford（计算机科学系，牛津大学）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Accepted to COLM 2025

2504.08525 2025-08-26 cs.AI cs.CL 86%

Task Memory Engine (TME): Enhancing State Awareness for Multi-Step LLM Agent Tasks

Ye Ye

机构 * Ye Ye（独立研究者）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 14 pages, 5 figures. Preprint prepared for future submission. Includes implementation and token-efficiency analysis. Code at https://github.com/biubiutomato/TME-Agent

2508.04903 2025-08-13 cs.CL cs.AI cs.MA 86%

RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory

Jun Liu, Zhenglun Kong, Changdi Yang, Fan Yang, Tianqi Li, Peiyan Dong, Joannah Nanjekye, Hao Tang, Geng Yuan, Wei Niu, Wenbin Zhang, Pu Zhao, Xue Lin, Dong Huang, Yanzhi Wang

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2508.00305 2025-08-04 cs.CL cs.LG cs.PF 86%

Systematic Evaluation of Optimization Techniques for Long-Context Language Models

Ammar Ahmed, Sheng Di, Franck Cappello, Zirui Liu, Jingoo Han, Ali Anwar

机构 * University of Minnesota（明尼苏达大学）； Argonne National Laboratory（阿贡国家实验室）； Samsung Semiconductor Inc.（三星半导体公司）

专题命中长上下文与记忆：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.CL、cs.LG

2507.22925 2025-08-01 cs.CL cs.AI 86%

Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents

Haoran Sun, Shaoning Zeng

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2507.19353 2025-07-28 cs.CL cs.AI 86%

Smooth Reading: Bridging the Gap of Recurrent LLM to Self-Attention LLM on Long-Context Tasks

Kai Liu, Zhan Su, Peijie Dong, Fengran Mo, Jianfei Gao, ShaoTing Zhang, Kai Chen

机构 * Shanghai AI Laboratory（上海人工智能实验室）； Tongji University（同济大学）； University of Montreal（蒙特利尔大学）； HKUST(GZ)（香港科技大学（广州））

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2507.05330 2025-07-09 cs.CL cs.AI 86%

MindFlow: Revolutionizing E-commerce Customer Support with Multimodal LLM Agents

Ming Gong, Xucheng Huang, Chenghan Yang, Xianhan Peng, Haoxin Wang, Yang Liu, Ling Jiang

机构 * Xiaoduo AI（小多AI）； University of Dayton（代顿大学）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2507.04625 2025-07-08 cs.CL cs.AI 86%

Knowledge-Aware Self-Correction in Language Models via Structured Memory Graphs

Swayamjit Saha

专题命中长上下文与记忆：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.CL、cs.AI

Comments 8 pages, 4 figures

2507.03042 2025-07-08 cs.CL cs.AI 86%

Dynamic Long Short-Term Memory Based Memory Storage For Long Horizon LLM Interaction

Yuyang Lou, Charles Li

机构 * University of Washington（华盛顿大学）

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 7 pages, 4 figures, 2 tables

2506.14606 2025-06-18 cs.CL cs.AR cs.LG cs.PL cs.SE 86%

Guaranteed Guess: A Language Modeling Approach for CISC-to-RISC Transpilation with Testing Guarantees

Ahmed Heakl, Sarim Hashmi, Chaimaa Abi, Celine Lee, Abdulrahman Mahmoud

机构 * MBZUAI（人工智能研究所）； Cornell University（康奈尔大学）

专题命中长上下文与记忆：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.CL、cs.LG

Comments Project page: https://ahmedheakl.github.io/Guaranteed-Guess/

2412.13879 2025-05-27 cs.CL cs.AI cs.CR 86%

Crabs: Consuming Resource via Auto-generation for LLM-DoS Attack under Black-box Settings

Yuanhe Zhang, Zhenhong Zhou, Wei Zhang, Xinyue Wang, Xiaojun Jia, Yang Liu, Sen Su

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 22 pages, 8 figures, 11 tables

2408.16978 2025-05-14 cs.DC cs.AI cs.LG 86%

Training Ultra Long Context Language Model with Fully Pipelined Distributed Transformer

Jinghan Yao, Sam Ade Jacobs, Masahiro Tanaka, Olatunji Ruwase, Hari Subramoni, Dhabaleswar K. Panda

专题命中长上下文与记忆：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.AI、cs.LG

Comments The Eighth Annual Conference on Machine Learning and Systems (MLSys'25)

2502.09017 2025-04-09 cs.CL cs.LG 86%

Diversity Enhances an LLM's Performance in RAG and Long-context Task

Zhichao Wang, Bin Bi, Yanqi Luo, Sitaram Asur, Claire Na Cheng

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

2503.17994 2025-03-25 cs.CL cs.AI 86%

Instructing the Architecture Search for Spatial-temporal Sequence Forecasting with LLM

Xin Xue, Haoyi Zhou, Tianyu Chen, Shuai Zhang, Yizhou Long, Jianxin Li

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2503.16071 2025-03-21 cs.CL cs.AI cs.IR 86%

Tuning LLMs by RAG Principles: Towards LLM-native Memory

Jiale Wei, Shuchi Wu, Ruochen Liu, Xiang Ying, Jingbo Shang, Fangbo Tao

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2503.07903 2025-03-12 cs.CL cs.LG 86%

Can Memory-Augmented Language Models Generalize on Reasoning-in-a-Haystack Tasks?

Payel Das, Ching-Yun Ko, Sihui Dai, Georgios Kollias, Subhajit Chaudhury, Aurelie Lozano

专题命中长上下文与记忆：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.CL、cs.LG

2503.04874 2025-03-10 cs.CL cs.LG 86%

Memory Is All You Need: Testing How Model Memory Affects LLM Performance in Annotation Tasks

Joan C. Timoneda, Sebastián Vallejo Vera

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

2503.02628 2025-03-05 cs.CL cs.AI 86%

Towards Event Extraction with Massive Types: LLM-based Collaborative Annotation and Partitioning Extraction

Wenxuan Liu, Zixuan Li, Long Bai, Yuxin Zuo, Daozhu Xu, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Work in progress

2502.15007 2025-02-24 cs.CL cs.AI 86%

LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers

Anton Razzhigaev, Matvey Mikhalchuk, Temurbek Rahmatullaev, Elizaveta Goncharova, Polina Druzhinina, Ivan Oseledets, Andrey Kuznetsov

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments accepted to NAACL 2025

2412.13148 2025-02-24 cs.LG cs.AI 86%

SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training

Chao Ma, Wenbo Gong, Meyer Scetbon, Edward Meeds

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments In v2 we have revised the related work, added more comprehensive citations, and clarified our key contributions

2406.06773 2025-02-14 cs.CL cs.AI 86%

Evaluating Zero-Shot Long-Context LLM Compression

Chenyu Wang, Yihan Wang, Kai Li

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2406.05925 2025-02-14 cs.CL cs.AI 86%

Hello Again! LLM-powered Personalized Agent for Long-term Dialogue

Hao Li, Chenghao Yang, An Zhang, Yang Deng, Xiang Wang, Tat-Seng Chua

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Accepted to NAACL 2025

2502.08910 2025-02-14 cs.CL cs.LG 86%

InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU

Heejun Lee, Geon Park, Jaduk Suh, Sung Ju Hwang

专题命中长上下文与记忆：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.CL、cs.LG

Comments 21 pages

2502.08586 2025-02-13 cs.LG cs.AI 86%

Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks

Ang Li, Yin Zhou, Vethavikashini Chithrra Raghuram, Tom Goldstein, Micah Goldblum

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

2412.15266 2024-12-23 cs.CL cs.AI 86%

On the Structural Memory of LLM Agents

Ruihong Zeng, Jinyuan Fang, Siwei Liu, Zaiqiao Meng

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2412.05896 2024-12-10 cs.LG cs.CL 86%

XKV: Personalized KV Cache Memory Reduction for Long-Context LLM Inference

Weizhuo Li, Zhigang Wang, Yu Gu, Ge Yu

专题命中长上下文与记忆：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

1. 长上下文与记忆 4760 篇

Beamforming-LLM: What, Where and When Did I Miss?

Banishing LLM Hallucinations Requires Rethinking Generalization

LLM-Assisted Iterative Evolution with Swarm Intelligence Toward SuperBrain

ByteScale: Efficient Scaling of LLM Training with a 2048K Context Length on More Than 12,000 GPUs

CRABS: A syntactic-semantic pincer strategy for bounding LLM interpretation of Python notebooks

Task Memory Engine (TME): Enhancing State Awareness for Multi-Step LLM Agent Tasks

RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory

Systematic Evaluation of Optimization Techniques for Long-Context Language Models

Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents

Smooth Reading: Bridging the Gap of Recurrent LLM to Self-Attention LLM on Long-Context Tasks

MindFlow: Revolutionizing E-commerce Customer Support with Multimodal LLM Agents

Knowledge-Aware Self-Correction in Language Models via Structured Memory Graphs

Dynamic Long Short-Term Memory Based Memory Storage For Long Horizon LLM Interaction

Guaranteed Guess: A Language Modeling Approach for CISC-to-RISC Transpilation with Testing Guarantees

Crabs: Consuming Resource via Auto-generation for LLM-DoS Attack under Black-box Settings

Training Ultra Long Context Language Model with Fully Pipelined Distributed Transformer

Diversity Enhances an LLM's Performance in RAG and Long-context Task

Instructing the Architecture Search for Spatial-temporal Sequence Forecasting with LLM

Tuning LLMs by RAG Principles: Towards LLM-native Memory

Can Memory-Augmented Language Models Generalize on Reasoning-in-a-Haystack Tasks?

Memory Is All You Need: Testing How Model Memory Affects LLM Performance in Annotation Tasks

Towards Event Extraction with Massive Types: LLM-based Collaborative Annotation and Partitioning Extraction

LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers

SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training

Evaluating Zero-Shot Long-Context LLM Compression

Hello Again! LLM-powered Personalized Agent for Long-term Dialogue

InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU

Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks

On the Structural Memory of LLM Agents

XKV: Personalized KV Cache Memory Reduction for Long-Context LLM Inference