语言大模型 / LLM - arXivDaily 专题

2209.11486 2023-02-06 cs.CL 70%

MetaPrompting: Learning to Learn Better Prompts

Yutai Hou, Hongyuan Dong, Xinghao Wang, Bohan Li, Wanxiang Che

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

Comments Accepted as COLING 2022 long paper

2301.02330 2023-01-09 cs.AI cs.HC q-bio.QM 70%

Evidence of behavior consistent with self-interest and altruism in an artificially intelligent agent

Tim Johnson, Nick Obradovich

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

2212.08354 2022-12-19 cs.CL 70%

FewFedWeight: Few-shot Federated Learning Framework across Multiple NLP Tasks

Weilong Dong, Xinwei Wu, Junzhuo Li, Shuangzhi Wu, Chao Bian, Deyi Xiong

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

2212.05856 2022-12-13 cs.CL 70%

"I think this is the most disruptive technology": Exploring Sentiments of ChatGPT Early Adopters using Twitter Data

Mubin Ul Haque, Isuru Dharmadasa, Zarrin Tasnim Sworna, Roshan Namal Rajapakse, Hussain Ahmad

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

Comments This is an early version of this paper

2206.03865 2022-12-13 cs.PL cs.AI cs.SE 70%

Fault-Aware Neural Code Rankers

Jeevana Priya Inala, Chenglong Wang, Mei Yang, Andres Codas, Mark Encarnación, Shuvendu K Lahiri, Madanlal Musuvathi, Jianfeng Gao

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

Comments In the proceedings of Advances in Neural Information Processing Systems, 2022

2208.11640 2022-12-06 cs.SE cs.AI cs.PL 70%

Repair Is Nearly Generation: Multilingual Program Repair with LLMs

Harshit Joshi, José Cambronero, Sumit Gulwani, Vu Le, Ivan Radicek, Gust Verbruggen

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

Comments 13 pages, Accepted at AAAI 2023

2210.08207 2022-11-18 cs.CL 70%

Temporal Word Meaning Disambiguation using TimeLMs

Mihir Godbole, Parth Dandavate, Aditya Kane

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

2207.04237 2022-09-09 cs.SE cs.LG 70%

Few-shot training LLMs for project-specific code-summarization

Toufique Ahmed, Premkumar Devanbu

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

Comments Accepted at ASE-NIER (2022) track

2209.03118 2022-09-08 cs.CL cs.CY 70%

The Ethical Need for Watermarks in Machine-Generated Language

Alexei Grinbaum, Laurynas Adomaitis

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

2208.12097 2022-08-26 cs.CL 70%

Training a T5 Using Lab-sized Resources

Manuel R. Ciosici, Leon Derczynski

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

2205.11601 2022-05-25 cs.CL cs.CY 70%

Challenges in Measuring Bias via Open-Ended Language Generation

Afra Feyza Akyürek, Muhammed Yusuf Kocyigit, Sejin Paik, Derry Wijaya

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

Comments 4th Workshop on Gender Bias in Natural Language Processing. NAACL, 2022

2205.09744 2022-05-20 cs.LG cs.CY cs.MM 70%

Overcoming Language Disparity in Online Content Classification with Multimodal Learning

Gaurav Verma, Rohit Mujumdar, Zijie J. Wang, Munmun De Choudhury, Srijan Kumar

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

Comments Accepted for publication at ICWSM 2022 as a full paper

2112.08674 2022-05-06 cs.CL 70%

Reframing Human-AI Collaboration for Generating Free-Text Explanations

Sarah Wiegreffe, Jack Hessel, Swabha Swayamdipta, Mark Riedl, Yejin Choi

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

Comments NAACL 2022 Camera-ready. 13 pages main + references, 14 pages appendix

2103.12407 2022-03-25 cs.CL 70%

Detecting Hate Speech with GPT-3

Ke-Li Chiu, Annie Collins, Rohan Alexander

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

Comments 29 pages, 1 figure, 23 tables 24 March 2022: Re-submission changes the modelling to occur multiple times and adds standard errors

2203.05648 2022-03-14 cs.CL 70%

Contextualized Sensorimotor Norms: multi-dimensional measures of sensorimotor strength for ambiguous English words, in context

Sean Trott, Benjamin Bergen

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

2202.02635 2022-02-08 cs.CL 70%

Multilingual Hate Speech and Offensive Content Detection using Modified Cross-entropy Loss

Arka Mitra, Priyanshu Sankhala

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

2201.10066 2022-01-26 cs.CL cs.DB 70%

Documenting Geographically and Contextually Diverse Data Sources: The BigScience Catalogue of Language Data and Resources

Angelina McMillan-Major, Zaid Alyafeai, Stella Biderman, Kimbo Chen, Francesco De Toni, Gérard Dupont, Hady Elsahar, Chris Emezue, Alham Fikri Aji, Suzana Ilić, Nurulaqilla Khamis, Colin Leong, Maraim Masoud, Aitor Soroa, Pedro Ortiz Suarez, Zeerak Talat, Daniel van Strien, Yacine Jernite

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

Comments 8 pages plus appendix and references

2109.06513 2022-01-17 cs.CL 70%

Exploring Prompt-based Few-shot Learning for Grounded Dialog Generation

Chujie Zheng, Minlie Huang

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

Comments Work in progress

2201.03382 2022-01-11 cs.CL 70%

BERT for Sentiment Analysis: Pre-trained and Fine-Tuned Alternatives

Frederico Souza, João Filho

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

Comments 10 pages, 1 figure, 3 tables. Accepted at International Conference on the Computational Processing of Portuguese (PROPOR 2022), but not yet published

2106.07890 2021-11-19 math.CT cs.CL 70%

An enriched category theory of language: from syntax to semantics

Tai-Danae Bradley, John Terilla, Yiannis Vlassopoulos

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

Comments 29 pages; v2 major revision with new proofs and computations

2111.02878 2021-11-05 cs.CL cs.IR 70%

Unsupervised and Distributional Detection of Machine-Generated Text

Matthias Gallé, Jos Rozen, Germán Kruszewski, Hady Elsahar

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

Comments 10 pages

2109.14728 2021-10-01 cs.HC cs.AI 70%

Collaborative Storytelling with Human Actors and AI Narrators

Boyd Branch, Piotr Mirowski, Kory W. Mathewson

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

Comments 5 pages, 1 figure, accepted to ICCC as Short Paper: Event Report

2106.10619 2021-06-22 cs.CL 70%

A Brief Study on the Effects of Training Generative Dialogue Models with a Semantic loss

Prasanna Parthasarathi, Mohamed Abdelsalam, Joelle Pineau, Sarath Chandar

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

Comments Accepted at SIGDial 2021

1903.11437 2019-03-28 cs.CL 70%

Using Monolingual Data in Neural Machine Translation: a Systematic Study

Franck Burlot, François Yvon

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

Comments Published in the Proceedings of the Third Conference on Machine Translation (Research Papers), 2018

2511.02622 2026-05-20 q-bio.BM physics.bio-ph physics.comp-ph 69%

Machine Learning for RNA Secondary Structure Prediction: a review of current methods and challenges

利用机器学习进行RNA二级结构预测：当前方法和挑战的综述

Giuseppe Sacco, Giovanni Bussi, Guido Sanguinetti

专题命中其他LLM ：foundation model(abstract,comments)；prompting(abstract)

AI总结本文综述了RNA二级结构预测中当前方法和挑战，重点讨论了机器学习和深度学习在该领域的应用，以及数据稀缺性带来的泛化危机，同时展望了未来需要解决的难题，如复杂结构的预测和动态结构的建模。

Comments 22 pages, 3 figures. Updated version of the article published in RNA 32(4):443-456 (2026); the Foundation Models section has been revised to reflect developments since publication. The remainder of the manuscript is unchanged apart from formatting

Journal ref RNA 32(4): 443-456 (2026)

2608.27321 2026-08-28 math.DS math.CA 新提交 67%

A blueprint for the formalization of norm-variation of multiple ergodic averages for commuting transformations

关于交换变换多重遍历平均的范数变差形式化的蓝图

Floris van Doorn, Polona Durcik, Joris Roos, Lenka Slavíková, Christoph Thiele

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

AI总结该研究以蓝图形式为Lean 4中交换变换多重遍历平均的范数变差结果的形式化提供基础，利用前沿大语言模型完成形式化，强化了陶哲轩的定理并回答了开放问题。

Comments 116 pages; associated formalization available at https://github.com/roos-j/lean-nct

2608.23688 2026-08-26 astro-ph.SR astro-ph.IM 新提交 67%

Agentic Active Learning Meets Visual Embeddings: Finding Anomalies among 370 000 Variable Stars from ASAS-SN

智能体主动学习结合视觉嵌入：从ASAS-SN的370000颗变星中发现异常天体

Milan Pesta, Yuan-Sen Ting

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

AI总结该研究提出结合多模态大语言模型智能体的主动学习框架，从ASAS-SN的37万余颗变星中高效发现异常天体，仅用约3小时、约40美元成本便得到含24个新异常的星表，验证了该方法的可行性。

Comments 24 pages, 12 figures, 6 tables. Submitted to ApJ

2608.23551 2026-08-25 cs.CL cs.AI cs.LG stat.ML 新提交 67%

ConvergeFlow: Language Flow with Provable Convergence to Token Embeddings

ConvergeFlow：可收敛至词元嵌入的语言流模型

Na Li, Yuchen Jiao, Changxiao Cai, Gen Li

机构 * Chinese University of Hong Kong（香港中文大学）； University of Michigan（密歇根大学）

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

AI总结本研究提出ConvergeFlow，一种嵌入空间的基于流的语言模型，通过约束数据预测器至词元嵌入凸包实现可收敛性，无需交叉熵解码器，在OpenWebText上取得与现有模型相当的性能，展现了基于流范式的语言建模潜力。

2608.23550 2026-08-25 cs.HC cs.CR 新提交 67%

When "Do Not" Is Not Deny: Security Rules in CLAUDE.md vs Built-In Controls

当“不要”并非“拒绝”：CLAUDE.md中的安全规则与内置控制

Ting Yan

专题命中其他LLM ：LLM(abstract,abstract_cn)

AI总结该研究对比CLAUDE.md的自然语言安全规则与Claude Code内置控制的匹配度，发现仅4%-16%的规则有匹配控制，揭示开发者无法获知规则是否被强制执行的安全问题。

Comments 11 pages, 5 figures

2608.21483 2026-08-25 cs.SE cs.PL 新提交 67%

SLICE: Specification-Level Isolation of Contract Enforcement

SLICE：规约级别的契约执行隔离

Soohan Lim, Hyundong Jin, Yo-Sub Han

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

AI总结 SLICE是一个分阶段的代码生成框架，通过规约结构化、函数体生成、契约断言生成三个阶段，在ContractEval数据集上使代码满足功能需求与输入契约的性能平均提升6.58%。

Comments 17 pages, 6 figures, 3 tables

1. 其他LLM 12266 篇

MetaPrompting: Learning to Learn Better Prompts

Evidence of behavior consistent with self-interest and altruism in an artificially intelligent agent

FewFedWeight: Few-shot Federated Learning Framework across Multiple NLP Tasks

"I think this is the most disruptive technology": Exploring Sentiments of ChatGPT Early Adopters using Twitter Data

Fault-Aware Neural Code Rankers

Repair Is Nearly Generation: Multilingual Program Repair with LLMs

Temporal Word Meaning Disambiguation using TimeLMs

Few-shot training LLMs for project-specific code-summarization

The Ethical Need for Watermarks in Machine-Generated Language

Training a T5 Using Lab-sized Resources

Challenges in Measuring Bias via Open-Ended Language Generation

Overcoming Language Disparity in Online Content Classification with Multimodal Learning

Reframing Human-AI Collaboration for Generating Free-Text Explanations

Detecting Hate Speech with GPT-3

Contextualized Sensorimotor Norms: multi-dimensional measures of sensorimotor strength for ambiguous English words, in context

Multilingual Hate Speech and Offensive Content Detection using Modified Cross-entropy Loss

Documenting Geographically and Contextually Diverse Data Sources: The BigScience Catalogue of Language Data and Resources

Exploring Prompt-based Few-shot Learning for Grounded Dialog Generation

BERT for Sentiment Analysis: Pre-trained and Fine-Tuned Alternatives

An enriched category theory of language: from syntax to semantics

Unsupervised and Distributional Detection of Machine-Generated Text

Collaborative Storytelling with Human Actors and AI Narrators

A Brief Study on the Effects of Training Generative Dialogue Models with a Semantic loss

Using Monolingual Data in Neural Machine Translation: a Systematic Study

Machine Learning for RNA Secondary Structure Prediction: a review of current methods and challenges

A blueprint for the formalization of norm-variation of multiple ergodic averages for commuting transformations

Agentic Active Learning Meets Visual Embeddings: Finding Anomalies among 370 000 Variable Stars from ASAS-SN

ConvergeFlow: Language Flow with Provable Convergence to Token Embeddings

When "Do Not" Is Not Deny: Security Rules in CLAUDE.md vs Built-In Controls

SLICE: Specification-Level Isolation of Contract Enforcement