语言大模型 / LLM - arXivDaily 专题

2505.21969 2025-09-29 cs.RO cs.AI 57%

DORAEMON: Decentralized Ontology-aware Reliable Agent with Enhanced Memory Oriented Navigation

Tianjun Gu, Linfeng Li, Xuhong Wang, Chenghua Gong, Jingyu Gong, Zhizhong Zhang, Yuan Xie, Lizhuang Ma, Xin Tan

专题命中长上下文与记忆：language model(abstract)；分类 cs.AI

2509.19640 2025-09-25 cs.CL 57%

AutoSpec: An Agentic Framework for Automatically Drafting Patent Specification

Ryan Shea, Zhou Yu

机构 * Columbia University（哥伦比亚大学）

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

Comments EMNLP Findings 2025

2505.01475 2025-09-23 cs.SE cs.AI 57%

CodeSSM: Towards State Space Models for Code Understanding

Shweta Verma, Abhinav Anand, Mira Mezini

机构 * TU Darmstadt（图宾根大学）； Hessian Center for Artificial Intelligence（黑森人工智能中心）； National Research Center for Applied Cybersecurity ATHENE（应用网络安全国家研究中心ATHENE）

专题命中长上下文与记忆：pretraining(abstract)；分类 cs.AI

2509.11973 2025-09-16 cs.AI cs.MM cs.SD 57%

MusicSwarm: Biologically Inspired Intelligence for Music Composition

Markus J. Buehler

专题命中长上下文与记忆：foundation model(abstract)；分类 cs.AI

2509.11225 2025-09-16 cs.RO cs.AI 57%

MEMBOT: Memory-Based Robot in Intermittent POMDP

Youzhi Liang, Eyan Noronha

机构 * Department of Computer Science Stanford University（计算机科学系斯坦福大学）

专题命中长上下文与记忆：pretraining(abstract)；分类 cs.AI

2504.05312 2025-09-12 cs.IR cs.AI 57%

Towards Adaptive Memory-Based Optimization for Enhanced Retrieval-Augmented Generation

Qitao Qin, Yucong Luo, Yihang Lu, Zhibo Chu, Xiaoman Liu, Xianwei Meng

机构 * University of Science and Technology of China（中国科学技术大学）； Hefei Institutes of Physical Science, Chinese Academy of Sciences（中国科学院合肥研究院）

专题命中长上下文与记忆：language model(abstract)；分类 cs.AI

Comments Accept by ACL 2025 findings

2509.05298 2025-09-09 cs.HC cs.AI cs.MM 57%

Livia: An Emotion-Aware AR Companion Powered by Modular AI Agents and Progressive Memory Compression

Rui Xi, Xianghan Wang

专题命中长上下文与记忆：prompting(abstract)；分类 cs.AI

Comments Accepted to the Proceedings of the 2025 International Conference on Artificial Intelligence and Virtual Reality (AIVR 2025). \c{opyright} 2025 Springer. This is the author-accepted manuscript. Rui Xi and Xianghan Wang contributed equally to this work. The final version will be available via SpringerLink

2509.00293 2025-09-03 cs.DB cs.LG 57%

Illuminating Patterns of Divergence: DataDios SmartDiff for Large-Scale Data Difference Analysis

Aryan Poduri, Yashwant Tailor

机构 * DataDios

专题命中长上下文与记忆：LLM(abstract)；分类 cs.LG

Comments 10 pages, 4 figures

2107.06925 2025-08-22 cs.DC cs.LG 57%

Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines

Shigang Li, Torsten Hoefler

机构 * Department of Computer Science, ETH Zurich（计算机科学系，苏黎世联邦理工学院）

专题命中长上下文与记忆：language model(abstract)；分类 cs.LG

Comments Published in Proceedings of the 2021 International Conference for High Performance Computing, Networking, Storage and Analysis (SC'21), November 2021, Article No.: 27, Pages 1-14. Best Paper Finalist

2109.04939 2025-08-20 cs.CL 57%

Modeling Human Sentence Processing with Left-Corner Recurrent Neural Network Grammars

Ryo Yoshida, Hiroshi Noji, Yohei Oseki

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

Comments Accepted by EMNLP 2021

2508.10160 2025-08-15 cs.HC cs.LG 57%

Pre-trained Transformer-models using chronic invasive electrophysiology for symptom decoding without patient-individual training

Timon Merk, Saeed Salehi, Richard M. Koehler, Qiming Cui, Maria Olaru, Amelia Hahn, Nicole R. Provenza, Simon Little, Reza Abbasi-Asl, Phil A. Starr, Wolf-Julian Neumann

专题命中长上下文与记忆：foundation model(abstract)；分类 cs.LG

Comments 5 pages, 6 figures

2507.04377 2025-08-12 cs.CV cs.CL cs.MM 57%

Multi-Modal Semantic Parsing for the Interpretation of Tombstone Inscriptions

Xiao Zhang, Johan Bos

机构 * University of Groningen（格罗宁根大学）

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

Comments ACMMM 2025

2504.12436 2025-08-12 cs.CV cs.AI 57%

Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation

Nairouz Mrabah, Nicolas Richet, Ismail Ben Ayed, Éric Granger

机构 * LIVIA ； ILLS ； Department of Systems Engineering, ÉTS Montreal（系统工程系，蒙特利尔ÉTS）； ÉTS Montreal（蒙特利尔ÉTS）

专题命中长上下文与记忆：language model(abstract)；分类 cs.AI

Comments ICCV2025

2409.01071 2025-08-05 cs.CV cs.CL 57%

VideoLLaMB: Long Streaming Video Understanding with Recurrent Memory Bridges

Yuxuan Wang, Yiqi Song, Cihang Xie, Yang Liu, Zilong Zheng

机构 * NLCo Lab, State Key Laboratory of General Artificial Intelligence, BIGAI（NLCo实验室、国家一般人工智能重点实验室、BIGAI）； School of Computer Science & Technology, Beijing Institute of Technology（计算机科学与技术学院、北京理工大学）； Computer Science and Engineering, University of California（计算机科学与工程系、加州大学）； Wangxuan Institute of Computer Technology, Peking University（王轩计算机技术研究所、北京大学）

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

Comments To appear at ICCV 2025

2504.14200 2025-07-31 cs.CV cs.AI 57%

Enhancing Multimodal In-Context Learning for Image Classification through Coreset Optimization

Huiyi Chen, Jiawei Peng, Kaihua Tang, Xin Geng, Xu Yang

机构 * Southeast University（东南大学）； Huawei Singapore Research Center（华为新加坡研究中心）

专题命中长上下文与记忆：language model(abstract)；分类 cs.AI

Comments 11 pages, 5 figures

2507.21378 2025-07-30 cs.HC cs.AI 57%

ProMemAssist: Exploring Timely Proactive Assistance Through Working Memory Modeling in Multi-Modal Wearable Devices

Kevin Pu, Ting Zhang, Naveen Sendhilnathan, Sebastian Freitag, Raj Sodhi, Tanya Jonker

机构 * University of Toronto（多伦多大学）； Meta Reality Labs（Meta现实实验室）

专题命中长上下文与记忆：LLM(abstract)；分类 cs.AI

Comments Accepted to UIST'25

2504.00927 2025-07-14 cs.CL 57%

Multi-Token Attention

Olga Golovneva, Tianlu Wang, Jason Weston, Sainbayar Sukhbaatar

机构 * FAIR at Meta（Meta 的 FAIR）

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

2507.06839 2025-07-10 cs.LG stat.ML 57%

Scalable Gaussian Processes: Advances in Iterative Methods and Pathwise Conditioning

Jihao Andreas Lin

专题命中长上下文与记忆：prompting(abstract)；分类 cs.LG

Comments PhD Thesis, University of Cambridge

2507.05753 2025-07-09 cs.LG 57%

Jigsaw: Training Multi-Billion-Parameter AI Weather Models with Optimized Model Parallelism

Deifilia Kieckhefen, Markus Götz, Lars H. Heyen, Achim Streit, Charlotte Debus

机构 * Karlsruhe Institute of Technology（卡尔斯鲁厄理工大学）； Karlsruhe Institute of Technology, Helmholtz AI（卡尔斯鲁厄理工大学、海姆霍兹人工智能）

专题命中长上下文与记忆：foundation model(abstract)；分类 cs.LG

Comments 12 pages, 10 figures

2410.07531 2025-07-08 cs.AR cs.AI 57%

Reducing the Cost of Dropout in Flash-Attention by Hiding RNG with GEMM

Haiyue Ma, Jian Liu, Ronny Krashinsky

机构 * Princeton University, NVIDIA（普林斯顿大学，NVIDIA）； NVIDIA

专题命中长上下文与记忆：LLM(abstract)；分类 cs.AI

2507.02001 2025-07-04 cs.LG 57%

Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames

Anurag Arnab, Ahmet Iscen, Mathilde Caron, Alireza Fathi, Cordelia Schmid

机构 * Google DeepMind（谷歌DeepMind）

专题命中长上下文与记忆：language model(abstract)；分类 cs.LG

2507.01652 2025-07-03 cs.CV cs.AI cs.MM 57%

Autoregressive Image Generation with Linear Complexity: A Spatial-Aware Decay Perspective

Yuxin Mao, Zhen Qin, Jinxing Zhou, Hui Deng, Xuyang Shen, Bin Fan, Jing Zhang, Yiran Zhong, Yuchao Dai

机构 * Northwestern Polytechnical University（西北工业大学）； TapTap ； OpenNLPLab ； Hefei University of Technology（合肥工业大学）； Australian National University（澳大利亚国立大学）

专题命中长上下文与记忆：language model(abstract)；分类 cs.AI

2507.00380 2025-07-02 cs.CL 57%

Gregorian melody, modality, and memory: Segmenting chant with Bayesian nonparametrics

Vojtěch Lanz, Jan Hajič

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

2506.24019 2025-07-01 cs.CV cs.CL 57%

Ella: Embodied Social Agents with Lifelong Memory

Hongxin Zhang, Zheyuan Zhang, Zeyuan Wang, Zunzhe Zhang, Lixing Fang, Qinhong Zhou, Chuang Gan

机构 * University of Massachusetts Amherst（马萨诸塞大学阿姆赫斯特分校）； Johns Hopkins University（约翰霍普金斯大学）； Tsinghua University（清华大学）

专题命中长上下文与记忆：foundation model(abstract)；分类 cs.CL

2506.17121 2025-06-23 cs.CL 57%

Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs?

Adithya Bhaskar, Alexander Wettig, Tianyu Gao, Yihe Dong, Danqi Chen

机构 * Department of Computer Science（计算机科学系）； Cranberry-Lemon University（Cranberry-Lemon 大学）； Department of Computational Neuroscience（计算神经科学系）； University of the Witwatersrand（沃斯兰德大学）

专题命中长上下文与记忆：language model(abstract)；分类 cs.CL

Comments We release our code publicly at https://github.com/princeton-pli/PruLong

2506.14035 2025-06-18 cs.CV cs.AI 57%

SimpleDoc: Multi-Modal Document Understanding with Dual-Cue Page Retrieval and Iterative Refinement

Chelsi Jain, Yiran Wu, Yifan Zeng, Jiale Liu, S hengyu Dai, Zhenwen Shao, Qingyun Wu, Huazheng Wang

机构 * Oregon State University（俄勒冈州立大学）； Pennsylvania State University（宾夕法尼亚州立大学）； AG2AI, Inc.（AG2AI公司）； Johnson & Johnson（强生公司）

专题命中长上下文与记忆：language model(abstract)；分类 cs.AI

2506.06310 2025-06-10 eess.SP cs.LG eess.IV 57%

Enhancing Contrastive Learning-based Electrocardiogram Pretrained Model with Patient Memory Queue

Xiaoyu Sun, Yang Yang, Xunde Dong

机构 * School of Automation Science and Engineering, South China University of Technology, Guangzhou, China（南方科技大学自动化科学与工程学院）

专题命中长上下文与记忆：pretraining(abstract)；分类 cs.LG

Comments 8 pages, 4 figures

2506.05370 2025-06-09 cs.AI cs.ET 57%

Contextual Memory Intelligence -- A Foundational Paradigm for Human-AI Collaboration and Reflective Generative AI Systems

Kristy Wedel

机构 * Franklin University（弗兰克林大学）

专题命中长上下文与记忆：LLM(abstract)；分类 cs.AI

Comments 32 pages, 9 tables, 1 figure

2505.03320 2025-06-04 cs.CL 57%

Recall with Reasoning: Chain-of-Thought Distillation for Mamba's Long-Context Memory and Extrapolation

Junyu Ma, Tianqing Fang, Zhisong Zhang, Hongming Zhang, Haitao Mi, Dong Yu

机构 * Tencent AI Lab（腾讯AI实验室）

专题命中长上下文与记忆：pretraining(abstract)；分类 cs.CL

2411.04999 2025-05-30 cs.RO cs.LG 57%

DynaMem: Online Dynamic Spatio-Semantic Memory for Open World Mobile Manipulation

Peiqi Liu, Zhanqiu Guo, Mohit Warke, Soumith Chintala, Chris Paxton, Nur Muhammad Mahi Shafiullah, Lerrel Pinto

机构 * New York University（纽约大学）； Meta Inc.（Meta公司）； Hello Robot Inc.（Hello Robot公司）

专题命中长上下文与记忆：language model(abstract)；分类 cs.LG

Comments Website: https://dynamem.github.io

1. 长上下文与记忆 4789 篇

DORAEMON: Decentralized Ontology-aware Reliable Agent with Enhanced Memory Oriented Navigation

AutoSpec: An Agentic Framework for Automatically Drafting Patent Specification

CodeSSM: Towards State Space Models for Code Understanding

MusicSwarm: Biologically Inspired Intelligence for Music Composition

MEMBOT: Memory-Based Robot in Intermittent POMDP

Towards Adaptive Memory-Based Optimization for Enhanced Retrieval-Augmented Generation

Livia: An Emotion-Aware AR Companion Powered by Modular AI Agents and Progressive Memory Compression

Illuminating Patterns of Divergence: DataDios SmartDiff for Large-Scale Data Difference Analysis

Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines

Modeling Human Sentence Processing with Left-Corner Recurrent Neural Network Grammars

Pre-trained Transformer-models using chronic invasive electrophysiology for symptom decoding without patient-individual training

Multi-Modal Semantic Parsing for the Interpretation of Tombstone Inscriptions

Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation

VideoLLaMB: Long Streaming Video Understanding with Recurrent Memory Bridges

Enhancing Multimodal In-Context Learning for Image Classification through Coreset Optimization

ProMemAssist: Exploring Timely Proactive Assistance Through Working Memory Modeling in Multi-Modal Wearable Devices

Multi-Token Attention

Scalable Gaussian Processes: Advances in Iterative Methods and Pathwise Conditioning

Jigsaw: Training Multi-Billion-Parameter AI Weather Models with Optimized Model Parallelism

Reducing the Cost of Dropout in Flash-Attention by Hiding RNG with GEMM

Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames

Autoregressive Image Generation with Linear Complexity: A Spatial-Aware Decay Perspective

Gregorian melody, modality, and memory: Segmenting chant with Bayesian nonparametrics

Ella: Embodied Social Agents with Lifelong Memory

Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs?

SimpleDoc: Multi-Modal Document Understanding with Dual-Cue Page Retrieval and Iterative Refinement

Enhancing Contrastive Learning-based Electrocardiogram Pretrained Model with Patient Memory Queue

Contextual Memory Intelligence -- A Foundational Paradigm for Human-AI Collaboration and Reflective Generative AI Systems

Recall with Reasoning: Chain-of-Thought Distillation for Mamba's Long-Context Memory and Extrapolation

DynaMem: Online Dynamic Spatio-Semantic Memory for Open World Mobile Manipulation