语言大模型 / LLM - arXivDaily 专题

2510.22629 2025-10-28 cs.CL cs.AI 79%

Integrating Linguistics and AI: Morphological Analysis and Corpus development of Endangered Toto Language of West Bengal

Ambalika Guha, Sajal Saha, Debanjan Ballav, Soumi Mitra, Hritwick Chakraborty

专题命中预训练与数据：language model(abstract)；small language model(abstract)；SLM(abstract)；分类 cs.CL、cs.AI

2502.04465 2025-10-28 cs.LG cs.AI cs.SD eess.AS 79%

FocalCodec: Low-Bitrate Speech Coding via Focal Modulation Networks

Luca Della Libera, Francesco Paissan, Cem Subakan, Mirco Ravanelli

机构 * Concordia University（康科德大学）； Mila-Quebec AI Institute（魁北克AI研究院）； Fondazione Bruno Kessler（布鲁诺·凯斯勒基金会）； University of Trento（特伦托大学）； Université Laval（拉瓦尔大学）

专题命中预训练与数据：large language model(abstract)；language model(abstract)；pretraining(abstract)；分类 cs.AI、cs.LG

Comments Accepted at NeurIPS 2025

2509.23041 2025-10-27 cs.CR cs.AI cs.CL 79%

Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data

Zi Liang, Qingqing Ye, Xuan Liu, Yanyun Wang, Jianliang Xu, Haibo Hu

机构 * The Hong Kong Polytechnic University（香港理工大学）； University of California, San Diego（加州大学圣地亚哥分校）； The Hong Kong University of Science and Technology (Guangzhou)（香港科学与技术大学（广州））； Hong Kong Baptist University（香港 Baptist 大学）； PolyU Research Centre for Privacy and Security Technologies in Future Smart Systems（PolyU 未来智能系统隐私与安全技术研究中心）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Camera Ready of NeurIPS 2025 Spotlight. Source code: https://github.com/liangzid/VirusInfectionAttack

2510.20154 2025-10-24 cs.CL cs.AI 79%

Are Stereotypes Leading LLMs' Zero-Shot Stance Detection ?

Anthony Dubreuil, Antoine Gourru, Christine Largeron, Amine Trabelsi

机构 * Laboratoire Hubert Curien, UMR CNRS 5516（Hubert Curien实验室，CNRS UMR 5516）； Department of Computer Science, Université de Sherbrooke（计算机科学系，Sherbrooke大学）

专题命中预训练与数据：large language model(abstract)；language model(abstract)；pretraining(abstract)；分类 cs.CL、cs.AI

Comments Accepted in EMNLP 2025 (Main)

2510.15851 2025-10-20 cs.CL cs.LG 79%

SpeechLLMs for Large-scale Contextualized Zero-shot Slot Filling

Kadri Hacioglu, Manjunath K E, Andreas Stolcke

机构 * Uniphore

专题命中预训练与数据：large language model(abstract)；language model(abstract)；foundation model(abstract)；分类 cs.CL、cs.LG

Comments 13 pages, EMNLP 2025

2504.00339 2025-10-14 cs.CL cs.AI 79%

VNJPTranslate: A comprehensive pipeline for Vietnamese-Japanese translation

Hoang Hai Phan, Nguyen Duc Minh Vu, Nam Dang Phuong

专题命中预训练与数据：large language model(abstract)；language model(abstract)；prompting(abstract)；分类 cs.CL、cs.AI

Comments The paper contains a critical error in Section 3.1, leading to invalid results in Section 3.3. This undermines the main conclusion of the paper. The authors are working on a corrected version, but in the meantime, there is not a quick fix/replacement/update available

2510.08744 2025-10-13 cs.LG cs.AI 79%

Graph Diffusion Transformers are In-Context Molecular Designers

Gang Liu, Jie Chen, Yihan Zhu, Michael Sun, Tengfei Luo, Nitesh V Chawla, Meng Jiang

机构 * University of Notre Dame（诺丁汉大学）； MIT-IBM Watson AI Lab, IBM Research（麻省理工-IBM Watson人工智能实验室，IBM研究）

专题命中预训练与数据：language model(abstract)；foundation model(abstract)；pretraining(abstract)；分类 cs.AI、cs.LG

Comments 29 pages, 16 figures, 17 tables. Model available at: https://huggingface.co/liuganghuggingface/DemoDiff-0.7B

2510.06250 2025-10-13 cs.CL cs.AI 79%

Scalable multilingual PII annotation for responsible AI in LLMs

Bharti Meena, Joanna Skubisz, Harshit Rajgarhia, Nand Dave, Kiran Ganesh, Shivali Dalmia, Abhishek Mukherji, Vasudevan Sundarababu

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2510.06151 2025-10-08 cs.LG cs.AI cs.HC 79%

LLMs as Policy-Agnostic Teammates: A Case Study in Human Proxy Design for Heterogeneous Agent Teams

Aju Ani Justus, Chris Baber

机构 * School of Computer Science, University of Birmingham（伯明翰大学计算机科学学院）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments This is a preprint of a paper presented at the \textit{European Conference on Artificial Intelligence (ECAI 2025)}. It is made publicly available for the benefit of the research community and should be regarded as a preprint rather than a formally reviewed publication

2510.04031 2025-10-07 cs.CL cs.AI 79%

Does Using Counterfactual Help LLMs Explain Textual Importance in Classification?

Nelvin Tan, James Asikin Cheung, Yu-Ching Shih, Dong Yang, Amol Salunkhe

机构 * American Express（美国运通）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 8 pages, 2 figures

2508.16048 2025-10-07 cs.CL cs.AI 79%

OpenWHO: A Document-Level Parallel Corpus for Health Translation in Low-Resource Languages

Raphaël Merx, Hanna Suominen, Trevor Cohn, Ekaterina Vylomova

机构 * School of Computing and Information Systems, The University of Melbourne（墨尔本大学计算与信息系统学院）； School of Computing, The Australian National University（澳大利亚国立大学计算学院）； School of Medicine and Psychology, The Australian National University（澳大利亚国立大学医学与心理学学院）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Accepted at WMT 2025

2508.01918 2025-10-06 cs.CL cs.AI 79%

Quantum-RAG and PunGPT2: Advancing Low-Resource Language Generation and Retrieval for the Punjabi Language

Jaskaranjeet Singh, Rakesh Thakur

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2510.00347 2025-10-02 cs.LG cs.AI cs.MA 79%

In-Context Curiosity: Distilling Exploration for Decision-Pretrained Transformers on Bandit Tasks

Huitao Yang, Guanting Chen

机构 * University of California, Los Angeles（加州大学洛杉矶分校）； University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）

专题命中预训练与数据：large language model(abstract)；language model(abstract)；pretraining(abstract)；分类 cs.AI、cs.LG

2505.11939 2025-09-30 eess.SP cs.AI cs.LG 79%

Fine-grained Contrastive Learning for ECG-Report Alignment with Waveform Enhancement

Haitao Li, Che Liu, Zhengyao Ding, Ziyi Liu, Wenqi Shao, Zhengxing Huang

机构 * Zhejiang University（浙江大学）； Imperial College London（伦敦帝国学院）； Transtek Medical Electronics Co., Ltd（Transtek医疗电子有限公司）； Shanghai AI Lab（上海AI实验室）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

2509.16197 2025-09-22 cs.CV cs.CL cs.LG 79%

MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer

Yanghao Li, Rui Qian, Bowen Pan, Haotian Zhang, Haoshuo Huang, Bowen Zhang, Jialing Tong, Haoxuan You, Xianzhi Du, Zhe Gan, Hyunjik Kim, Chao Jia, Zhenbang Wang, Yinfei Yang, Mingfei Gao, Zi-Yi Dou, Wenze Hu, Chang Gao, Dongxu Li, Philipp Dufter, Zirui Wang, Guoli Yin, Zhengdong Zhang, Chen Chen, Yang Zhao, Ruoming Pang, Zhifeng Chen

机构 * Apple（苹果公司）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

2503.22723 2025-09-19 cs.LG cs.AI 79%

Zero-Shot LLMs in Human-in-the-Loop RL: Replacing Human Feedback for Reward Shaping

Mohammad Saif Nazir, Chayan Banerjee

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

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments 20 pages, 3 figures, 4 Tables

2508.00033 2025-09-17 cs.SE cs.AI cs.CL 79%

GPT-4.1 Sets the Standard in Automated Experiment Design Using Novel Python Libraries

Nuno Fachada, Daniel Fernandes, Carlos M. Fernandes, Bruno D. Ferreira-Saraiva, João P. Matos-Carvalho

机构 * Copelabs, Lusófona University（Copelabs与卢斯奥法纳大学）； Center of Technology and Systems (UNINOVA-CTS) and Associated Lab of Intelligent Systems (LASI)（技术与系统中心（UNINOVA-CTS）及智能系统关联实验室）； CICANT, Lusófona University（CICANT与卢斯奥法纳大学）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments The peer-reviewed version of this paper is published in Future Internet at https://doi.org/10.3390/fi17090412. This version is typeset by the author and differs only in pagination and typographical detail

Journal ref Future Internet. 2025; 17(9):412

2509.09174 2025-09-12 cs.CL cs.AI cs.SD 79%

EchoX: Towards Mitigating Acoustic-Semantic Gap via Echo Training for Speech-to-Speech LLMs

Yuhao Zhang, Yuhao Du, Zhanchen Dai, Xiangnan Ma, Kaiqi Kou, Benyou Wang, Haizhou Li

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2508.21788 2025-09-01 cs.CL cs.AI cs.IR 79%

Going over Fine Web with a Fine-Tooth Comb: Technical Report of Indexing Fine Web for Problematic Content Search and Retrieval

Inés Altemir Marinas, Anastasiia Kucherenko, Andrei Kucharavy

机构 * École Polytechnique Fédérale de Lausanne（瑞士联邦理工学院）； Institute of Entrepreneurship and Management, HES-SO Valais-Wallis（创业与管理研究所）； Institute of Informatics, HES-SO Valais-Wallis（信息研究所）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2508.21632 2025-09-01 cs.CL cs.AI 79%

QZhou-Embedding Technical Report

Peng Yu, En Xu, Bin Chen, Haibiao Chen, Yinfei Xu

机构 * Kingsoft AI（金山人工智能）

专题命中预训练与数据：LLM(abstract)；foundation model(abstract)；pretraining(abstract)；分类 cs.CL、cs.AI

2508.20236 2025-08-29 math.HO cs.AI cs.HC cs.LG 79%

The Mathematician's Assistant: Integrating AI into Research Practice

Jonas Henkel

机构 * Fachbereich Mathematik und Informatik, Philipps-Universität Marburg（数学与计算机科学系，马尔堡菲利普斯大学）

专题命中预训练与数据：large language model(abstract)；language model(abstract)；prompting(abstract)；分类 cs.AI、cs.LG

Comments 24 pages, 7 figures. Accepted for publication in Mathematische Semesterberichte (to appear in vol. 72, no. 2)

2508.18384 2025-08-27 cs.CL cs.AI 79%

Backprompting: Leveraging Synthetic Production Data for Health Advice Guardrails

Kellen Tan Cheng, Anna Lisa Gentile, Chad DeLuca, Guang-Jie Ren

机构 * IBM Research - Almaden（IBM研究院-阿玛登分部）； Princeton University（普林斯顿大学）； Adobe Research（Adobe研究院）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2508.18132 2025-08-26 cs.IR cs.AI cs.LG 79%

Test-Time Scaling Strategies for Generative Retrieval in Multimodal Conversational Recommendations

Hung-Chun Hsu, Yuan-Ching Kuo, Chao-Han Huck Yang, Szu-Wei Fu, Hanrong Ye, Hongxu Yin, Yu-Chiang Frank Wang, Ming-Feng Tsai, Chuan-Ju Wang

机构 * Research Center for Information Technology Innovation, Academia Sinica（资讯科技创新研究所以）； NVIDIA（NVIDIA公司）； Department of Computer Science, National Chengchi University（国立政治大学计算机科学系）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

2508.17057 2025-08-26 cs.CL cs.CR cs.LG 79%

GRAID: Synthetic Data Generation with Geometric Constraints and Multi-Agentic Reflection for Harmful Content Detection

Melissa Kazemi Rad, Alberto Purpura, Himanshu Kumar, Emily Chen, Mohammad Shahed Sorower

机构 * Capital One, AI Foundations（Capital One人工智能基础）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Comments 19 pages, 12 figures

2508.10000 2025-08-15 cs.CL cs.LG 79%

AutoGeTS: Knowledge-based Automated Generation of Text Synthetics for Improving Text Classification

Chenhao Xue, Yuanzhe Jin, Adrian Carrasco-Revilla, Joyraj Chakraborty, Min Chen

机构 * FabLab, Inetum（FabLab，Inetum）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

2508.06533 2025-08-12 cs.CL cs.AI 79%

The Art of Breaking Words: Rethinking Multilingual Tokenizer Design

Aamod Thakur, Ajay Nagpal, Atharva Savarkar, Kundeshwar Pundalik, Siddhesh Dosi, Piyush Sawarkar, Viraj Thakur, Rohit Saluja, Maunendra Sankar Desarkar, Ganesh Ramakrishnan

机构 * BharatGen Team（BharatGen团队）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2508.05954 2025-08-11 cs.CV cs.AI cs.CL 79%

Bifrost-1: Bridging Multimodal LLMs and Diffusion Models with Patch-level CLIP Latents

Han Lin, Jaemin Cho, Amir Zadeh, Chuan Li, Mohit Bansal

机构 * UNC Chapel Hill（北卡罗来纳大学教堂山分校）； Lambda

专题命中预训练与数据：large language model(abstract)；language model(abstract)；pretraining(abstract)；分类 cs.CL、cs.AI

Comments Project Page: https://bifrost-1.github.io

2508.04401 2025-08-07 cs.CL cs.AI 79%

Why are LLMs' abilities emergent?

Vladimír Havlík

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 20 pages

2508.02601 2025-08-05 cs.LG cs.AI 79%

StructSynth: Leveraging LLMs for Structure-Aware Tabular Data Synthesis in Low-Data Regimes

Siyi Liu, Yujia Zheng, Yongqi Zhang

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

2508.02209 2025-08-05 cs.LG cs.AI cs.IT cs.NI math.IT 79%

Balancing Information Accuracy and Response Timeliness in Networked LLMs

Yigit Turkmen, Baturalp Buyukates, Melih Bastopcu

机构 * School of Computer Science University of Birmingham Birmingham, UK（计算机科学学院布拉姆斯希尔大学布拉姆斯希尔英国）

专题命中预训练与数据：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

1. 预训练与数据 12393 篇

Integrating Linguistics and AI: Morphological Analysis and Corpus development of Endangered Toto Language of West Bengal

FocalCodec: Low-Bitrate Speech Coding via Focal Modulation Networks

Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data

Are Stereotypes Leading LLMs' Zero-Shot Stance Detection ?

SpeechLLMs for Large-scale Contextualized Zero-shot Slot Filling

VNJPTranslate: A comprehensive pipeline for Vietnamese-Japanese translation

Graph Diffusion Transformers are In-Context Molecular Designers

Scalable multilingual PII annotation for responsible AI in LLMs

LLMs as Policy-Agnostic Teammates: A Case Study in Human Proxy Design for Heterogeneous Agent Teams

Does Using Counterfactual Help LLMs Explain Textual Importance in Classification?

OpenWHO: A Document-Level Parallel Corpus for Health Translation in Low-Resource Languages

Quantum-RAG and PunGPT2: Advancing Low-Resource Language Generation and Retrieval for the Punjabi Language

In-Context Curiosity: Distilling Exploration for Decision-Pretrained Transformers on Bandit Tasks

Fine-grained Contrastive Learning for ECG-Report Alignment with Waveform Enhancement

MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer

Zero-Shot LLMs in Human-in-the-Loop RL: Replacing Human Feedback for Reward Shaping

GPT-4.1 Sets the Standard in Automated Experiment Design Using Novel Python Libraries

EchoX: Towards Mitigating Acoustic-Semantic Gap via Echo Training for Speech-to-Speech LLMs

Going over Fine Web with a Fine-Tooth Comb: Technical Report of Indexing Fine Web for Problematic Content Search and Retrieval

QZhou-Embedding Technical Report

The Mathematician's Assistant: Integrating AI into Research Practice

Backprompting: Leveraging Synthetic Production Data for Health Advice Guardrails

Test-Time Scaling Strategies for Generative Retrieval in Multimodal Conversational Recommendations

GRAID: Synthetic Data Generation with Geometric Constraints and Multi-Agentic Reflection for Harmful Content Detection

AutoGeTS: Knowledge-based Automated Generation of Text Synthetics for Improving Text Classification

The Art of Breaking Words: Rethinking Multilingual Tokenizer Design

Bifrost-1: Bridging Multimodal LLMs and Diffusion Models with Patch-level CLIP Latents

Why are LLMs' abilities emergent?

StructSynth: Leveraging LLMs for Structure-Aware Tabular Data Synthesis in Low-Data Regimes

Balancing Information Accuracy and Response Timeliness in Networked LLMs