语言大模型 / LLM - arXivDaily 专题

2509.15089 2025-09-19 cs.CL 92%

LLM-OREF: An Open Relation Extraction Framework Based on Large Language Models

Hongyao Tu, Liang Zhang, Yujie Lin, Xin Lin, Haibo Zhang, Long Zhang, Jinsong Su

机构 * School of Informatics, Xiamen University（厦门大学信息学院）； LLM Team, Shopee Pte. Ltd.（Shopee Pte. Ltd. 机器学习团队）； National Institute for Data Science in Health and Medicine, Xiamen University（厦门大学医学数据科学国家研究院）

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

2509.14526 2025-09-19 cs.CL cs.AI cs.LG 90%

Delta Knowledge Distillation for Large Language Models

Yihan Cao, Yanbin Kang, Zhengming Xing, Ruijie Jiang

机构 * LinkedIn Corporation（领英公司）

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

Comments 8 pages, 3 figures

2509.14930 2025-09-19 cs.CL cs.AI 90%

Cross-Modal Knowledge Distillation for Speech Large Language Models

Enzhi Wang, Qicheng Li, Zhiyuan Tang, Yuhang Jia

机构 * TMCC, College of Computer Science, Nankai University, Tianjin, China（TMCC，计算机科学学院，南开大学，天津，中国）； Tencent Ethereal Audio Lab, Tencent Corporation, Shenzhen, China（腾讯虚实音频实验室，腾讯公司，深圳，中国）

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

2509.14680 2025-09-19 cs.MA cs.LG 85%

LEED: A Highly Efficient and Scalable LLM-Empowered Expert Demonstrations Framework for Multi-Agent Reinforcement Learning

Tianyang Duan, Zongyuan Zhang, Songxiao Guo, Dong Huang, Yuanye Zhao, Zheng Lin, Zihan Fang, Dianxin Luan, Heming Cui, Yong Cui

机构 * Department of Computer Science, The University of Hong Kong（香港大学计算机科学系）； Institute of Data Science, National University of Singapore（新加坡国立大学数据科学研究所）； College of International Education, Hebei University of Economics and Business（河北经贸大学国际教育学院）； Department of Electrical and Electronic Engineering, The University of Hong Kong（香港大学电子与电气工程系）； Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系）； Institute for Imaging, Data and Communications, University of Edinburgh（爱丁堡大学成像、数据与通讯研究所）； Department of Computer Science and Technology, Tsinghua University（清华大学计算机科学与技术系）

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

Comments 5 pages, 4 figures

2509.01566 2025-09-19 cs.IR cs.CL 85%

CSRM-LLM: Embracing Multilingual LLMs for Cold-Start Relevance Matching in Emerging E-commerce Markets

Yujing Wang, Yiren Chen, Huoran Li, Chunxu Xu, Yuchong Luo, Xianghui Mao, Cong Li, Lun Du, Chunyang Ma, Qiqi Jiang, Yin Wang, Fan Gao, Wenting Mo, Pei Wen, Shantanu Kumar, Taejin Park, Yiwei Song, Vijay Rajaram, Tao Cheng, Sonu Durgia, Pranam Kolari

机构 * Coupang, Inc.（Coupang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

Comments 7 pages, 3 figures, accepted by CIKM 2025

2509.14781 2025-09-19 cs.AR 85%

LEAP: LLM Inference on Scalable PIM-NoC Architecture with Balanced Dataflow and Fine-Grained Parallelism

Yimin Wang, Yue Jiet Chong, Xuanyao Fong

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

Comments Accepted to the 2025 International Conference on Computer-Aided Design (ICCAD'25)

2509.15097 2025-09-19 cs.LG 81%

The Energy-Efficient Hierarchical Neural Network with Fast FPGA-Based Incremental Learning

Mohammad Saleh Vahdatpour, Huaiyuan Chu, Yanqing Zhang

机构 * Department of Computer Science Georgia State University Atlanta, GA 30302-5060, USA（计算机科学系佐治亚州立大学阿拉巴马州亚特兰大 30302-5060, 美国）

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

Comments Published at IJCNN 2025

2509.15076 2025-09-19 cs.LG cs.CV 79%

Forecasting and Visualizing Air Quality from Sky Images with Vision-Language Models

Mohammad Saleh Vahdatpour, Maryam Eyvazi, Yanqing Zhang

机构 * Georgia State University（佐治亚州立大学）； Savannah College of Art and Design（萨凡纳艺术与设计学院）

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

Comments Published at ICCVW 2025

2505.23024 2025-09-19 cs.LG 79%

An Empirical Study of Federated Prompt Learning for Vision Language Model

Zhihao Wang, Wenke Huang, Tian Chen, Zekun Shi, Guancheng Wan, Yu Qiao, Bin Yang, Jian Wang, Bing Li, Mang Ye

机构 * School of Computer Science, Wuhan University（武汉大学计算机学院）； Zhongguancun Laboratory, China（中关村实验室）

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

2509.13775 2025-09-19 cs.CL cs.AI 79%

Exploring Data and Parameter Efficient Strategies for Arabic Dialect Identifications

Vani Kanjirangat, Ljiljana Dolamic, Fabio Rinaldi

机构 * SUPSI（瑞士苏黎世联邦理工学院）； IDSIA（智能系统研究所）； armasuisse S+T

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

Comments 4 main pages, 4 additional, 5 figures

2509.14689 2025-09-19 cs.CL 74%

HARNESS: Lightweight Distilled Arabic Speech Foundation Models

Vrunda N. sukhadia, Shammur Absar Chowdhury

机构 * Qatar Computing Research Institute（卡塔尔计算研究院）

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

Comments 5 pages, 4 figures

2509.14886 2025-09-19 cs.CL cs.AI 73%

A Multi-To-One Interview Paradigm for Efficient MLLM Evaluation

Ye Shen, Junying Wang, Farong Wen, Yijin Guo, Qi Jia, Zicheng Zhang, Guangtao Zhai

机构 * Shanghai Jiao Tong University（上海交通大学）； Shanghai AI Laboratory（上海人工智能实验室）； Fudan University（复旦大学）

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

Comments 5 pages, 2 figures

2509.14547 2025-09-19 cs.AI 70%

(P)rior(D)yna(F)low: A Priori Dynamic Workflow Construction via Multi-Agent Collaboration

Yi Lin, Lujin Zhao, Yijie Shi

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

2509.13723 2025-09-19 cs.CL 70%

DSPC: Dual-Stage Progressive Compression Framework for Efficient Long-Context Reasoning

Yaxin Gao, Yao Lu, Zongfei Zhang, Jiaqi Nie, Shanqing Yu, Qi Xuan

机构 * Institute of Cyberspace Security, Zhejiang University of Technology A STAR Amazon Binjiang Institute of Artificial Intelligence, Zhejiang University of Technology

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

2509.14626 2025-09-19 cs.SE 67%

Evaluating the Effectiveness of Coverage-Guided Fuzzing for Testing Deep Learning Library APIs

Feiran Qin, M. M. Abid Naziri, Hengyu Ai, Saikat Dutta, Marcelo d'Amorim

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

2509.15211 2025-09-19 cs.CL 57%

What's the Best Way to Retrieve Slides? A Comparative Study of Multimodal, Caption-Based, and Hybrid Retrieval Techniques

Petros Stylianos Giouroukis, Dimitris Dimitriadis, Dimitrios Papadopoulos, Zhenwen Shao, Grigorios Tsoumakas

机构 * Aristotle University of Thessaloniki（亚里士多德大学）

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

2509.15113 2025-09-19 cs.LG 57%

Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers

Andrei Chertkov, Artem Basharin, Mikhail Saygin, Evgeny Frolov, Stanislav Straupe, Ivan Oseledets

机构 * Artificial Intelligence Research Institute (AIRI)（人工智能研究 institute）； Skolkovo Institute of Science and Technology（斯克尔科沃科学与技术研究所）； Sber Quantum Technology Center（Sber 量子技术中心）

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

2509.15038 2025-09-19 cs.CL 57%

Value-Guided KV Compression for LLMs via Approximated CUR Decomposition

Ayan Sengupta, Siddhant Chaudhary, Tanmoy Chakraborty

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

2506.02279 2025-09-19 cs.CL 57%

ImpRAG: Retrieval-Augmented Generation with Implicit Queries

Wenzheng Zhang, Xi Victoria Lin, Karl Stratos, Wen-tau Yih, Mingda Chen

机构 * Rutgers University（新泽西罗格斯大学）； FAIR, Meta（FAIR，Meta）

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

Comments Findings of EMNLP 2025

2509.15224 2025-09-19 cs.CV 50%

Depth AnyEvent: A Cross-Modal Distillation Paradigm for Event-Based Monocular Depth Estimation

Luca Bartolomei, Enrico Mannocci, Fabio Tosi, Matteo Poggi, Stefano Mattoccia

机构 * Advanced Research Center on Electronic System (ARCES)（电子系统先进研究中心）； Department of Computer Science and Engineering (DISI)（计算机科学与工程系）； University of Bologna（博洛尼亚大学）

专题命中效率与部署：foundation model(abstract)

Comments ICCV 2025. Code: https://github.com/bartn8/depthanyevent/ Project Page: https://bartn8.github.io/depthanyevent/

2509.14737 2025-09-19 cs.SD 50%

Pushing the Limits of End-to-End Diarization

Samuel J. Broughton, Lahiru Samarakoon

机构 * FanoHong Kong SAR, China（Fano香港特别行政区，中国）

专题命中效率与部署：pretraining(abstract)

Comments As presented at Interspeech 2025

Journal ref In Proc. Interspeech 2025 (pp. 5218-5222)

2509.14527 2025-09-19 cs.MM cs.SD 50%

CLAIP-Emo: Parameter-Efficient Adaptation of Language-supervised models for In-the-Wild Audiovisual Emotion Recognition

Yin Chen, Jia Li, Jinpeng Hu, Zhenzhen Hu, Richang Hong

机构 * Hefei University of Technology（合肥工业大学）

专题命中效率与部署：foundation model(abstract)

Comments The code and models will be available at https://github.com/MSA-LMC/CLAIP-Emo

2406.05951 2025-09-19 cs.RO 50%

Open-Vocabulary Part-Based Grasping

Tjeard van Oort, Dimity Miller, Will N. Browne, Nicolas Marticorena, Jesse Haviland, Niko Suenderhauf

机构 * Queensland University of Technology（昆士兰理工大学）

专题命中效率与部署：foundation model(abstract)

1. 效率与部署 23 篇

LLM-OREF: An Open Relation Extraction Framework Based on Large Language Models

Delta Knowledge Distillation for Large Language Models

Cross-Modal Knowledge Distillation for Speech Large Language Models

LEED: A Highly Efficient and Scalable LLM-Empowered Expert Demonstrations Framework for Multi-Agent Reinforcement Learning

CSRM-LLM: Embracing Multilingual LLMs for Cold-Start Relevance Matching in Emerging E-commerce Markets

LEAP: LLM Inference on Scalable PIM-NoC Architecture with Balanced Dataflow and Fine-Grained Parallelism

The Energy-Efficient Hierarchical Neural Network with Fast FPGA-Based Incremental Learning

Forecasting and Visualizing Air Quality from Sky Images with Vision-Language Models

An Empirical Study of Federated Prompt Learning for Vision Language Model

Exploring Data and Parameter Efficient Strategies for Arabic Dialect Identifications

HARNESS: Lightweight Distilled Arabic Speech Foundation Models

A Multi-To-One Interview Paradigm for Efficient MLLM Evaluation

(P)rior(D)yna(F)low: A Priori Dynamic Workflow Construction via Multi-Agent Collaboration

DSPC: Dual-Stage Progressive Compression Framework for Efficient Long-Context Reasoning

Evaluating the Effectiveness of Coverage-Guided Fuzzing for Testing Deep Learning Library APIs

What's the Best Way to Retrieve Slides? A Comparative Study of Multimodal, Caption-Based, and Hybrid Retrieval Techniques

Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers

Value-Guided KV Compression for LLMs via Approximated CUR Decomposition

ImpRAG: Retrieval-Augmented Generation with Implicit Queries

Depth AnyEvent: A Cross-Modal Distillation Paradigm for Event-Based Monocular Depth Estimation

Pushing the Limits of End-to-End Diarization

CLAIP-Emo: Parameter-Efficient Adaptation of Language-supervised models for In-the-Wild Audiovisual Emotion Recognition

Open-Vocabulary Part-Based Grasping