语言大模型 / LLM - arXivDaily 专题

2307.01446 2023-07-06 cs.CL cs.LG 88%

On Conditional and Compositional Language Model Differentiable Prompting

Jonathan Pilault, Can Liu, Mohit Bansal, Markus Dreyer

专题命中其他LLM ：language model(title,abstract)；prompting(title,abstract)；分类 cs.CL、cs.LG

Comments Accepted at International Joint Conference on Artificial Intelligence (IJCAI) 2023

2306.07875 2023-06-14 cs.IR cs.AI cs.CL cs.HC 88%

ReadProbe: A Demo of Retrieval-Enhanced Large Language Models to Support Lateral Reading

Dake Zhang, Ronak Pradeep

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

2306.07377 2023-06-14 cs.CL cs.AI 88%

Lost in Translation: Large Language Models in Non-English Content Analysis

Gabriel Nicholas, Aliya Bhatia

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

Comments 50 pages, 4 figures

2210.15458 2023-06-02 cs.CL cs.LG stat.ML 88%

Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models

Luke Vilnis, Yury Zemlyanskiy, Patrick Murray, Alexandre Passos, Sumit Sanghai

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.LG

Comments 17 pages, to appear at ICML 2023

2209.01515 2023-06-02 cs.CL cs.AI 88%

Do Large Language Models know what humans know?

Sean Trott, Cameron Jones, Tyler Chang, James Michaelov, Benjamin Bergen

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

2305.11778 2023-05-22 cs.CL cs.LG 88%

Cross-Lingual Supervision improves Large Language Models Pre-training

Andrea Schioppa, Xavier Garcia, Orhan Firat

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.LG

2304.01238 2023-05-09 cs.CL cs.AI 88%

Spam-T5: Benchmarking Large Language Models for Few-Shot Email Spam Detection

Maxime Labonne, Sean Moran

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

2303.17511 2023-03-31 cs.CY cs.AI cs.CL 88%

On pitfalls (and advantages) of sophisticated large language models

Anna Strasser

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

2303.10131 2023-03-20 cs.SE cs.AI cs.CY cs.LG 88%

She Elicits Requirements and He Tests: Software Engineering Gender Bias in Large Language Models

Christoph Treude, Hideaki Hata

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI、cs.LG

Comments 6 pages, MSR 2023

2303.00077 2023-03-02 cs.CL cs.AI 88%

Beyond the limitations of any imaginable mechanism: large language models and psycholinguistics

Conor Houghton, Nina Kazanina, Priyanka Sukumaran

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

Comments This is a commentary on Bowers Et. Al. (2023) doi:10.1017/S0140525X22002813

2212.09271 2022-12-21 cs.DB cs.AI cs.LG 88%

Very Large Language Model as a Unified Methodology of Text Mining

Meng Jiang

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI、cs.LG

Comments 4 pages, 3 figures

2510.27190 2025-11-03 cs.CR cs.AI 88%

Unvalidated Trust: Cross-Stage Vulnerabilities in Large Language Model Architectures

Dominik Schwarz

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI；LLM(comments)

Comments 178 pages, mechanism-centered taxonomy of 41 LLM risk patterns, extensive appendix with experiment prompts and consolidation tables. Full traces available to reviewers and affected providers

2510.21958 2025-10-28 cs.CL cs.DL 88%

A Stylometric Application of Large Language Models

Harrison F. Stropkay, Jiayi Chen, Mohammad J. Latifi, Daniel N. Rockmore, Jeremy R. Manning

机构 * Dartmouth College（达特茅斯学院）

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL；LLM(comments)

Comments All code and data needed to reproduce the results in this paper are available at https://github.com/ContextLab/llm-stylometry

2404.07546 2024-07-24 cs.CL 88%

Does In-Context Learning Really Learn? Rethinking How Large Language Models Respond and Solve Tasks via In-Context Learning

Quanyu Long, Yin Wu, Wenya Wang, Sinno Jialin Pan

专题命中其他LLM ：language model(title,abstract)；large language model(title,abstract)；分类 cs.CL

Comments 39 pages, 8 figures. Accepted by Conference On Language Modeling (COLM) 2024

2405.05080 2024-05-09 cs.HC cs.AI 88%

Concerns on Bias in Large Language Models when Creating Synthetic Personae

Helena A. Haxvig

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI；LLM(comments)

Comments 4 pages, accepted at the "LLM-Based Synthetic Personae and Data in HCI" workshop at CHI2024

2403.08882 2024-03-15 cs.MA cs.AI q-bio.PE 88%

Cultural evolution in populations of Large Language Models

Jérémy Perez, Corentin Léger, Marcela Ovando-Tellez, Chris Foulon, Joan Dussauld, Pierre-Yves Oudeyer, Clément Moulin-Frier

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI；LLM(comments)

Comments 17 pages, 20 figures. Open-source code available at https://github.com/jeremyperez2/LLM-Culture

2608.04375 2026-08-06 cs.CR 新提交 88%

Large Language Models and Social Media Information Integrity: Opportunities, Challenges, and Research Directions

大语言模型与社交媒体信息完整性：机遇、挑战与研究方向

Junjie Xiong, Zhengyuan Jiang, Xiaoran Xu, Chi Zhang, Changjia Zhu, Ning Wang, Mingkui Wei, Zhuo Lu, Yao Liu, Lingyao Li

专题命中其他LLM ：large language model(title,abstract)；language model(title,abstract)

AI总结本综述分析LLMs对社交媒体信息完整性的双重影响，指出其可增强恶意内容检测能力但也会生成欺骗性内容，提出跨语言检测等研究方向以利用LLMs并降低风险。

Comments It has been accepted by Computing Surveys. Preview From: htong@illinois.edu Congratulations! Your manuscript, "Large Language Models and Social Media Information Integrity: Opportunities, Challenges, and Research Directions," has been accepted for publication in ACM Computing Surveys.Your paper will be returned to your Author Center. Dr. Hanghang Tong Editor-in-Chief ACM Computing Surveys

Journal ref Just accpeted by ACM Computing Surveys 2026

2607.28331 2026-07-31 cs.SE 新提交 88%

Structural Validation of LLM-Generated Microservice Decompositions Using Source-Code Dependencies

基于源代码依赖的大语言模型生成微服务分解的结构验证

Daniel Silva, Renan Alves, Emanuel Dantas Filho, Ademar Sousa Neto, Mirko Perkusich, Danyllo Wagner Albuquerque, Kyller Gorgônio, Angelo Perkusich

专题命中其他LLM ：LLM(title,abstract)；large language model(abstract)；language model(abstract)；prompting(abstract)

AI总结本文提出基于静态依赖分析的自动验证流程，评估OpenAI o3生成的微服务分解的结构一致性，发现零样本与少样本提示的结构一致性相当，且需控制映射覆盖率以避免方法学偏差。

2604.19114 2026-07-13 cs.HC 88%

OOPrompt: Reifying Intents into Structured Artifacts for Modular and Iterative Prompting

OOPrompt: 将意图重构为结构化 artifact 以实现模块化和迭代提示

Tengyou Xu, Detao Ma, Xiang 'Anthony' Chen

专题命中其他LLM ：prompting(title,abstract)；LLM(abstract,abstract_cn)；large language model(abstract)；language model(abstract)

AI总结 OOPrompt 通过结构化 artifact 实现模块化和迭代提示，提升多意图提示的表达效率与可重用性。

Comments 20 pages, 8 figures, To appear in EICS 2026

Journal ref Proceedings of the ACM on Human-Computer Interaction, Volume 10, Issue 4, Article EICS009, 2026

2607.02808 2026-07-07 cs.SE 新提交 88%

A Systematic Methodology for Evaluating Failure Independence in LLM-Generated Code

一种评估大语言模型生成代码中故障独立性的系统方法

Rodrigo Pato Nogueira, Karthik Pattabiraman, Marco Vieira, João R. Campos

专题命中其他LLM ：LLM(title,abstract)；large language model(abstract)；language model(abstract)；prompting(abstract)

AI总结提出评估大语言模型生成代码中故障独立性的系统方法，应用于多模型、语言和提示策略的224个问题，通过多种分析得出大语言模型生成的解决方案不满足故障独立性假设，但异构模型有帮助，验证了该方法。

2605.31220 2026-06-25 cs.CL cs.AI cs.LG 版本更新 88%

Shared Doubt: Zero-Shot Cross-Lingual Confidence Estimation for Language Models

共享疑虑：语言模型的零样本跨语言置信度估计

Athina Kyriakou, Dennis Ulmer, Ivan Titov

机构 * ILLC, University of Amsterdam（阿姆斯特丹大学ILLC）； ILCC, University of Edinburgh（爱丁堡大学ILCC）

专题命中其他LLM ：language model(title,abstract)；LLM(abstract,abstract_cn)；large language model(abstract)；分类 cs.CL、cs.AI、cs.LG

AI总结研究多语言大语言模型是否编码共享的、可跨语言迁移的置信度特征，通过轻量级线性探针从中间表示直接预测答案正确性，实现零样本跨语言泛化，并发现置信度特征集中在中间层。

2606.17510 2026-06-18 cs.SE cs.SY eess.SY 新提交 88%

OmniDroneX: An LLM-Assisted Holistic Drone-as-a-Service Ecosystem

OmniDroneX: 一种LLM辅助的全方位无人机即服务生态系统

I-Ling Yen, Akeem Mohammed, Farokh Bastani, San-Yih Hwang

专题命中其他LLM ：LLM(title,title_cn)；large language model(abstract)；language model(abstract)

AI总结提出OmniDroneX统一无人机即服务生态系统，通过libUAV接口和PT-SOA抽象模型连接底层物理与高层任务，利用大语言模型辅助功能识别、服务组合和自然语言任务定义，支持多种组合技术以实现可扩展、自演进的无人机系统。

Comments This manuscript is a full version of a paper accepted in shortened form by IEEE International Conference on Joint Cloud Computing

2606.18120 2026-06-17 cs.CR cs.AI cs.CL cs.LG 新提交 88%

Structural Role Injection in Handlebars-Templated LLM Prompts: Triple-Brace Interpolation, Delimiter Family, and the Limits of HTML Auto-Escaping

Handlebars模板化LLM提示中的结构角色注入：三花括号插值、分隔符家族与HTML自动转义的局限性

Mohammadreza Rashidi

机构 * Department of Computer Science AI（计算机科学系人工智能）； Media Analysis Lab Berlin, Germany（媒体分析实验室柏林德国）

专题命中其他LLM ：LLM(title,title_cn)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI、cs.LG

AI总结本文研究Handlebars模板引擎中双花括号与三花括号插值对结构角色注入攻击的影响，通过无模型分析和5760次实验，揭示HTML转义仅保护特定分隔符家族，无法替代指令与数据的结构分离。

Comments 7 pages, 6 figures

2501.03957 2026-06-09 cs.HC cs.CV 88%

Vision Language Models as Values Detectors

视觉语言模型作为价值检测器

Giulio Antonio Abbo, Tony Belpaeme

机构 * IDLab-AIRO, Ghent University – imec, Belgium（IDLab-AIRO、根特大学 – imec、比利时）

专题命中其他LLM ：language model(title,abstract)；LLM(summary_cn,abstract_cn)；large language model(abstract)

AI总结本文研究了先进LLM与人类标注者在家庭环境场景中检测相关元素的对齐情况，发现LLaVA 34B表现最佳但仍需改进，表明LLM在检测图像中价值元素方面有潜力。

Comments 13 pages, 2 figures

Journal ref Value Engineering in Artificial Intelligence (VALE 2024) (LNAI,volume 15356)

2509.26464 2026-05-20 cs.AI cs.CL cs.LG 88%

Extreme Self-Preference in Language Models

语言模型中的极端自我偏好

Steven A. Lehr, Mary Cipperman, Mahzarin R. Banaji

机构 * Cangrade, Inc.（Cangrade公司）； Department of Physics, Harvard University（哈佛大学物理系）； Department of Psychology, Harvard University（哈佛大学心理学系）

专题命中其他LLM ：language model(title,abstract)；LLM(abstract,abstract_cn)；large language model(abstract)；分类 cs.CL、cs.AI、cs.LG

AI总结研究发现大型语言模型在字词关联任务中表现出对自身名称、公司和CEO的强烈偏好，这表明模型的自我认同可能影响其行为，引发对模型自我偏好影响的深入探讨。

Comments 73 pages total. Main article 22 pages, 6 main-text tables. Supplementary Materials (51 pages, 28 tables). Data, transcripts, and code for replication and data extraction have been uploaded to OSF: https://osf.io/98ye3/

2605.08896 2026-05-12 cs.CL cs.AI cs.LG 88%

FragileFlow: Spectral Control of Correct-but-Fragile Predictions for Foundation Model Robustness

FragileFlow: 对基础模型鲁棒性进行谱控制的正确但易碎预测的谱控制

Zhuoyun Li, Boxuan Wang, Jinwei Hu, Xiaowei Huang, Yi Dong

机构 * School of Computer Science and Informatics, University of Liverpool, UK（利兹大学计算机科学与信息学学院）

专题命中其他LLM ：LLM(summary_cn,abstract)；foundation model(title)；分类 cs.CL、cs.AI、cs.LG

AI总结本文提出FragileFlow，通过谱控制识别正确但易碎的预测，提升基础模型鲁棒性，实验显示其在多个选择LLM基准和少量样本CLIP适应中表现优异。

2510.26285 2026-04-23 cs.CL cs.AI cs.LG cs.NE 88%

Language Models Learn Universal Representations of Numbers and Here's Why You Should Care

语言模型学习通用数字表示及其原因

Michal Štefánik, Timothee Mickus, Marek Kadlčík, Bertram Højer, Michal Spiegel, Raúl Vázquez, Aman Sinha, Josef Kuchař, Philipp Mondorf, Pontus Stenetorp

机构 * R&D Centre for Large Language Models, National Institute of Informatics, Japan（日本信息处理国家研究所大型语言模型研发中心）； University of Helsinki（赫尔辛基大学）； IT University of Copenhagen（哥本哈根IT大学）； Kempelen Instutite of Information Technology（凯普莱恩信息科技研究所）； IECL-ATILF, Université de Lorraine - ICANS Strasbourg（洛林大学IECL-ATILF - 斯特拉斯堡ICANS）； MaiNLP, Center for Information and Language Processing, LMU Munich（慕尼黑技术大学MaiNLP，信息与语言处理中心）； Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心（MCML））； University College London（伦敦大学学院）

专题命中其他LLM ：language model(title,abstract)；LLM(abstract,abstract_cn)；large language model(abstract)；分类 cs.CL、cs.AI、cs.LG

AI总结研究揭示大语言模型在数字表示上的系统性与通用性，指出其对评估模型编码能力及减少算术错误的重要性。

2604.20803 2026-04-23 cs.SE 88%

Autonomous LLM-generated Feedback for Student Exercises in Introductory Software Engineering Courses

自主生成的LLM反馈用于初级软件工程课程学生练习

Andreas Metzger

专题命中其他LLM ：LLM(title,title_cn)

AI总结本文提出NAILA工具，通过LLM提供24/7自动反馈，解决初级软件工程课程中学生人数激增、背景多样及生成式AI影响带来的反馈挑战，通过实证研究评估其对学生学习效果的影响。

2312.06562 2026-03-17 cs.CL cs.AI cs.LG math.CT 88%

On Meta-Prompting

元提示

Adrian de Wynter, Xun Wang, Qilong Gu, Si-Qing Chen

专题命中其他LLM ：prompting(title,abstract)；LLM(abstract)；large language model(abstract)；language model(abstract)

AI总结本文基于范畴论提出理论框架，描述LLM与用户交互中的ICL行为，分析元提示在生成理想输出上的有效性。

Comments Preprint. Under review

2509.25369 2026-02-27 cs.CL cs.AI cs.LG 88%

Generative Value Conflicts Reveal LLM Priorities

生成价值冲突揭示大语言模型优先级

Andy Liu, Kshitish Ghate, Mona Diab, Daniel Fried, Atoosa Kasirzadeh, Max Kleiman-Weiner

机构 * Carnegie Mellon University（卡内基梅隆大学）； University of Washington（华盛顿大学）

专题命中其他LLM ：LLM(title,abstract)；large language model(abstract)；language model(abstract)；prompting(abstract)

AI总结生成价值冲突揭示大语言模型优先级，通过ConflictScope评估模型在价值冲突中的优先级，发现模型在开放式设置中更支持个人价值，系统提示能提高对齐效果14%。

Comments Accepted to ICLR 2026 (the 14th International Conference on Learning Representations)

1. 其他LLM 12145 篇

On Conditional and Compositional Language Model Differentiable Prompting

ReadProbe: A Demo of Retrieval-Enhanced Large Language Models to Support Lateral Reading

Lost in Translation: Large Language Models in Non-English Content Analysis

Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models

Do Large Language Models know what humans know?

Cross-Lingual Supervision improves Large Language Models Pre-training

Spam-T5: Benchmarking Large Language Models for Few-Shot Email Spam Detection

On pitfalls (and advantages) of sophisticated large language models

She Elicits Requirements and He Tests: Software Engineering Gender Bias in Large Language Models

Beyond the limitations of any imaginable mechanism: large language models and psycholinguistics

Very Large Language Model as a Unified Methodology of Text Mining

Unvalidated Trust: Cross-Stage Vulnerabilities in Large Language Model Architectures

A Stylometric Application of Large Language Models

Does In-Context Learning Really Learn? Rethinking How Large Language Models Respond and Solve Tasks via In-Context Learning

Concerns on Bias in Large Language Models when Creating Synthetic Personae

Cultural evolution in populations of Large Language Models

Large Language Models and Social Media Information Integrity: Opportunities, Challenges, and Research Directions

Structural Validation of LLM-Generated Microservice Decompositions Using Source-Code Dependencies

OOPrompt: Reifying Intents into Structured Artifacts for Modular and Iterative Prompting

A Systematic Methodology for Evaluating Failure Independence in LLM-Generated Code

Shared Doubt: Zero-Shot Cross-Lingual Confidence Estimation for Language Models

OmniDroneX: An LLM-Assisted Holistic Drone-as-a-Service Ecosystem

Structural Role Injection in Handlebars-Templated LLM Prompts: Triple-Brace Interpolation, Delimiter Family, and the Limits of HTML Auto-Escaping

Vision Language Models as Values Detectors

Extreme Self-Preference in Language Models

FragileFlow: Spectral Control of Correct-but-Fragile Predictions for Foundation Model Robustness

Language Models Learn Universal Representations of Numbers and Here's Why You Should Care

Autonomous LLM-generated Feedback for Student Exercises in Introductory Software Engineering Courses

On Meta-Prompting

Generative Value Conflicts Reveal LLM Priorities