语言大模型 / LLM - arXivDaily 专题

2406.10288 2025-03-03 cs.CL cs.LG 87%

Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models

Francisco Eiras, Aleksandar Petrov, Philip H. S. Torr, M. Pawan Kumar, Adel Bibi

专题命中指令微调：large language model(title)；language model(title)；prompting(abstract)；分类 cs.CL、cs.LG

Comments Accepted to ICLR'25

2502.12346 2025-02-19 cs.LG cs.AI 87%

QuZO: Quantized Zeroth-Order Fine-Tuning for Large Language Models

Jiajun Zhou, Yifan Yang, Kai Zhen, Ziyue Liu, Yequan Zhao, Ershad Banijamali, Athanasios Mouchtaris, Ngai Wong, Zheng Zhang

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.AI、cs.LG

2404.17985 2025-01-22 cs.CL cs.AI cs.SI 87%

Detection of Conspiracy Theories Beyond Keyword Bias in German-Language Telegram Using Large Language Models

Milena Pustet, Elisabeth Steffen, Helena Mihaljević

专题命中指令微调：large language model(title)；language model(title)；prompting(abstract)；分类 cs.CL、cs.AI

Comments Accepted to the 8th Workshop on Online Abuse and Harms (WOAH), ACL 2024

Journal ref in: 8th Workshop on Online Abuse and Harms (WOAH), ACL (2024)

2411.18242 2024-11-28 cs.CL cs.AI 87%

Thai Financial Domain Adaptation of THaLLE -- Technical Report

KBTG Labs, Atthakorn Petchsod, Pornchanan Balee, Danupat Khamnuansin, Anuruth Lertpiya, Chanatip Saetia, Tawunrat Chalothorn, Thadpong Pongthawornkamol, Monchai Lertsutthiwong

专题命中指令微调：LLM(abstract)；large language model(abstract)；language model(abstract)；pretraining(abstract)

2410.19025 2024-10-28 cs.LG cs.AI 87%

Large Language Models for Financial Aid in Financial Time-series Forecasting

Md Khairul Islam, Ayush Karmacharya, Timothy Sue, Judy Fox

专题命中指令微调：large language model(title)；language model(title)；foundation model(abstract)；分类 cs.AI、cs.LG

Comments GitHub link MLSys/Financial-Time-Series" target="_blank" rel="noopener">https://github.com/UVA-MLSys/Financial-Time-Series

2409.16799 2024-09-26 cs.AI cs.LG stat.AP 87%

Large Language Model Predicts Above Normal All India Summer Monsoon Rainfall in 2024

Ujjawal Sharma, Madhav Biyani, Akhil Dev Suresh, Debi Prasad Bhuyan, Saroj Kanta Mishra, Tanmoy Chakraborty

专题命中指令微调：large language model(title)；language model(title)；LLM(abstract)；分类 cs.AI、cs.LG

Comments 3 figures

2409.05385 2024-09-19 cs.CL cs.AI 87%

Towards Building a Robust Knowledge Intensive Question Answering Model with Large Language Models

Xingyun Hong, Yan Shao, Zhilin Wang, Manni Duan, Jin Xiongnan

专题命中指令微调：large language model(title)；language model(title)；LLM(abstract)；分类 cs.CL、cs.AI

Comments This paper has been accepted by NLPCC-2024

2409.00209 2024-09-04 cs.CL cs.LG 87%

Enhancing Event Reasoning in Large Language Models through Instruction Fine-Tuning with Semantic Causal Graphs

Mazal Bethany, Emet Bethany, Brandon Wherry, Cho-Yu Chiang, Nishant Vishwamitra, Anthony Rios, Peyman Najafirad

专题命中指令微调：large language model(title)；language model(title)；prompting(abstract)；分类 cs.CL、cs.LG

2408.16753 2024-08-30 cs.CL cs.LG 87%

Reinforcement Learning without Human Feedback for Last Mile Fine-Tuning of Large Language Models

Alec Solway

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.LG

2405.12884 2024-05-22 cs.CL cs.AI 87%

Investigating Persuasion Techniques in Arabic: An Empirical Study Leveraging Large Language Models

Abdurahmman Alzahrani, Eyad Babkier, Faisal Yanbaawi, Firas Yanbaawi, Hassan Alhuzali

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.AI

2401.07301 2024-05-14 cs.CL cs.AI 87%

Small Language Model Can Self-correct

Haixia Han, Jiaqing Liang, Jie Shi, Qianyu He, Yanghua Xiao

专题命中指令微调：language model(title,abstract)；small language model(title)；分类 cs.CL、cs.AI

2312.01552 2023-12-05 cs.CL cs.AI 87%

The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning

Bill Yuchen Lin, Abhilasha Ravichander, Ximing Lu, Nouha Dziri, Melanie Sclar, Khyathi Chandu, Chandra Bhagavatula, Yejin Choi

专题命中指令微调：LLM(abstract)；large language model(abstract)；language model(abstract)；SFT(abstract)

Comments 26 pages, 8 figures. Project website: https://allenai.github.io/re-align/

2311.18063 2023-12-01 cs.CL cs.LG cs.SI 87%

TurkishBERTweet: Fast and Reliable Large Language Model for Social Media Analysis

Ali Najafi, Onur Varol

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.LG

Comments 21 pages, 4 figures, 8 tables

2310.00035 2023-10-06 cs.LG cs.AI 87%

LoRA ensembles for large language model fine-tuning

Xi Wang, Laurence Aitchison, Maja Rudolph

专题命中指令微调：large language model(title)；language model(title)；LLM(abstract)；分类 cs.AI、cs.LG

Comments Update the title in the PDF file

2308.01240 2023-08-03 cs.CL cs.AI 87%

Evaluating Instruction-Tuned Large Language Models on Code Comprehension and Generation

Zhiqiang Yuan, Junwei Liu, Qiancheng Zi, Mingwei Liu, Xin Peng, Yiling Lou

专题命中指令微调：large language model(title)；language model(title)；instruction tuning(abstract)；分类 cs.CL、cs.AI

2303.10464 2023-08-01 cs.LG cs.CL 87%

SPDF: Sparse Pre-training and Dense Fine-tuning for Large Language Models

Vithursan Thangarasa, Abhay Gupta, William Marshall, Tianda Li, Kevin Leong, Dennis DeCoste, Sean Lie, Shreyas Saxena

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.LG

Comments Accepted to Uncertainty in Artificial Intelligence (UAI) 2023 Conference; 13 pages, 4 figures (Main Paper) + 5 pages (Supplementary Material)

2303.12132 2023-03-23 cs.CL cs.CR cs.LG 87%

Fundamentals of Generative Large Language Models and Perspectives in Cyber-Defense

Andrei Kucharavy, Zachary Schillaci, Loïc Maréchal, Maxime Würsch, Ljiljana Dolamic, Remi Sabonnadiere, Dimitri Percia David, Alain Mermoud, Vincent Lenders

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.LG

Comments 41 pages (without references), 13 figures; public report of Cyber-Defence Campus

2210.16637 2022-11-24 cs.CL cs.LG 87%

Beyond Prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering Representations

Yu Fei, Ping Nie, Zhao Meng, Roger Wattenhofer, Mrinmaya Sachan

专题命中指令微调：language model(title,abstract)；prompting(title)；分类 cs.CL、cs.LG

Comments Accepted to EMNLP 2022

2109.05687 2021-09-14 cs.CL cs.AI 87%

Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning

Runxin Xu, Fuli Luo, Zhiyuan Zhang, Chuanqi Tan, Baobao Chang, Songfang Huang, Fei Huang

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.AI

Comments Accepted as a long paper to EMNLP 2021 Main Conference

2102.00875 2021-02-02 cs.LG cs.CL cs.DC 87%

Scaling Federated Learning for Fine-tuning of Large Language Models

Agrin Hilmkil, Sebastian Callh, Matteo Barbieri, Leon René Sütfeld, Edvin Listo Zec, Olof Mogren

专题命中指令微调：language model(title,abstract)；large language model(title)；分类 cs.CL、cs.LG

2509.16275 2025-09-23 cs.CR cs.AI cs.SE 87%

SecureFixAgent: A Hybrid LLM Agent for Automated Python Static Vulnerability Repair

Jugal Gajjar, Kamalasankari Subramaniakuppusamy, Relsy Puthal, Kaustik Ranaware

机构 * Computer Science Department（计算机科学系）； Applied Economics Department（应用经济学系）

专题命中指令微调：LLM(title,abstract)；large language model(abstract,comments)；language model(abstract,comments)；分类 cs.AI

Comments 6 pages, 3 figures, 4 tables, 1 algorithm, accepted in the Robustness and Security of Large Language Models (ROSE-LLM) special session at ICMLA 2025

2401.09796 2024-01-22 cs.LG cs.CR 87%

A Fast, Performant, Secure Distributed Training Framework For Large Language Model

Wei Huang, Yinggui Wang, Anda Cheng, Aihui Zhou, Chaofan Yu, Lei Wang

专题命中指令微调：large language model(title)；language model(title)；LLM(abstract,comments)；分类 cs.LG

Comments Accepted by ICASSP 2024 (Federated LLM)

2501.12147 2026-08-18 cs.CL cs.AI cs.LG 版本更新 87%

Improving Influence-based Instruction Tuning Data Selection for Balanced Learning of Diverse Capabilities

改进基于影响力的指令调优数据选择以实现多样化能力的均衡学习

Qirun Dai, Dylan Zhang, Jiaqi W. Ma, Hao Peng

专题命中指令微调：instruction tuning(title)；LLM(abstract_cn)；large language model(abstract)；language model(abstract)

AI总结该研究针对基于影响力的指令调优数据选择的任务偏向问题，提出BIDS算法，经实验验证其在小比例数据下可实现更均衡性能且优于现有方法。

Comments Accepted to EMNLP 2025 (Findings)

2606.06027 2026-08-13 cs.AI cs.CL cs.LG cs.SI 版本更新 87%

RedditPersona: A Modular Framework for Community-Conditioned LLM Adaptation from Reddit

RedditPersona: 一个用于从Reddit进行社区条件化LLM适配的模块化框架

Amirhossein Ghaffari, Ali Goodarzi, Huong Nguyen, Simo Hosio, Lauri Lovén, Ekaterina Gilman

机构 * Future Computing Group University of Oulu（未来计算组奥卢大学）； Centre for Applied Computing University of Oulu（应用计算中心奥卢大学）

专题命中指令微调：LLM(title,title_cn)；language model(abstract)；分类 cs.CL、cs.AI、cs.LG

AI总结提出RedditPersona模块化框架，通过五种分组策略和QLoRA训练参数高效适配器，在112个Reddit子版块上评估社区条件化语言模型，发现适配器的行为可识别性与策略内在一致性相关，且所有策略在可识别性和分布相似性之间存在一致权衡。

Comments Accepted manuscript (CC BY 4.0). To appear in ACM CIKM 2026, Rome, Italy, Nov 7-11, 2026. DOI: https://doi.org/10.1145/3799682.3840182

2607.11528 2026-08-06 cs.CE 版本更新 87%

HermesHFL: Incentive-Compatible Hierarchical Federated Unlearning for Dynamic LLM Fine-Tuning

HermesHFL：用于动态大语言模型微调的激励兼容分层联邦遗忘

Chenxi Sun, Minghui Liwang, Wusi He, Yuhan Su, Zhang Liu, Sai Zou, Wei Ni, Seyyedali Hosseinalipour

专题命中指令微调：LLM(title,abstract)；large language model(abstract)；language model(abstract)

AI总结针对大语言模型微调中分层联邦遗忘面临的挑战，提出HermesHFL框架，通过参数高效微调支持选择性遗忘等。制定统一优化问题，开发Neogen框架解决，实验表明该框架在模型效用等多方面优于现有基线。

Comments 15pages,8 figures

2608.01759 2026-08-04 cs.CR 新提交 87%

Benign Alone, Harmful Together: Exploiting Experience Composition in Self-Evolving LLM Agents

单独无害，联合有害：在自进化大语言模型智能体中利用经验组合

Bingyu Yan, Xiaoming Zhang, Chaozhuo Li, Ziyi Zhou, Yirui Qi, Litian Zhang

专题命中指令微调：LLM(title,summary_cn)；large language model(abstract)；language model(abstract)

AI总结本研究提出EvoBreak攻击，利用自进化LLM智能体的良性经验组合，结合BreakGym生成训练目标，在保持高无害性的同时优于现有攻击，揭示了新的安全风险。

2608.00078 2026-08-04 cs.CV 新提交 87%

Device-First Feedback: Toward Mobile-Native LLM-Driven Neural Architecture Search

设备优先反馈：面向移动端原生大语言模型驱动的神经网络架构搜索

Saif U Din, Muhammad Ahsan Hussain, Radu Timofte, Dmitry Ignatov

机构 * University of Würzburg（维尔茨堡大学）； CAIDAS ； IFI

专题命中指令微调：LLM(title,abstract)；large language model(abstract)；language model(abstract)

AI总结该研究提出自动化移动端部署闭环流水线，针对CIFAR-10、CIFAR-100基准在三星平板验证，发现GPU微调无法保证移动端性能单调提升，需多数据集设备端测量。

Comments 11 pages, 4 figures, 4 tables. Code: https://github.com/ABrain-One/nn-gpt

2607.24232 2026-07-28 cs.IR 新提交 87%

Strategy-Aware Parameter-Efficient Adaptation for LLM-based Auto-Bidding

基于大语言模型的自动出价的策略感知参数高效适配

Songyue Cai, Lianyu Wang, Shan Gu, Ziru Xu, Jian Xu, Xiaofeng Zhu, Bo Zheng

专题命中指令微调：LLM(title,abstract)；large language model(abstract)；language model(abstract)

AI总结研究广告自动出价问题，提出SAGE框架，通过位置增强、文本对齐和约束门控LoRA三个组件，实现参数高效多模态对齐，在大规模基准实验中性能卓越，调整参数少，消融研究验证各组件贡献。

2509.05080 2026-07-28 q-fin.TR 版本更新 87%

MM-ARC: Multimodal Adaptive Routing of Capital with Robustness-Audited Strategy Pools

MM-ARC：具有稳健性审核策略池的资本多模态自适应路由

Yang Chen, Yuchen Cao, Jacky Keung, Leilei Gan, Kun Kuang, Yueheng Jiang, Zhaozhao Ma, Jianping Zhu, Fei Wu, Jinpeng Li

专题命中指令微调：LLM(title,abstract)；large language model(abstract)；language model(abstract)；SFT(abstract)

AI总结研究金融交易系统如何转化多模态市场历史，提出MM-ARC方法，通过多视图在不同专家间分配资本，经稳健性审核筛选候选方案，实验显示该方法在夏普比率和最大回撤等指标上优于基线，有相对基准的支持。

2602.06941 2026-07-07 cs.LG cs.AI cs.CL 版本更新 87%

Endogenous Resistance to Activation Steering in Language Models

语言模型中激活引导的内生抵抗

Alex McKenzie, Keenan Pepper, Stijn Servaes, Martin Leitgab, Murat Cubuktepe, Mike Vaiana, Diogo de Lucena, Judd Rosenblatt, Michael S. A. Graziano

机构 * University of Washington（华盛顿大学）

专题命中指令微调：language model(title,abstract)；large language model(abstract)；prompting(abstract)；分类 cs.CL、cs.AI、cs.LG

AI总结研究发现大型语言模型在任务不匹配的激活引导下能内生抵抗，通过显式重启恢复正确生成，并识别出相关稀疏自编码器潜在变量，可增强或削弱该抵抗。

1. 指令微调 11692 篇

Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models

QuZO: Quantized Zeroth-Order Fine-Tuning for Large Language Models

Detection of Conspiracy Theories Beyond Keyword Bias in German-Language Telegram Using Large Language Models

Thai Financial Domain Adaptation of THaLLE -- Technical Report

Large Language Models for Financial Aid in Financial Time-series Forecasting

Large Language Model Predicts Above Normal All India Summer Monsoon Rainfall in 2024

Towards Building a Robust Knowledge Intensive Question Answering Model with Large Language Models

Enhancing Event Reasoning in Large Language Models through Instruction Fine-Tuning with Semantic Causal Graphs

Reinforcement Learning without Human Feedback for Last Mile Fine-Tuning of Large Language Models

Investigating Persuasion Techniques in Arabic: An Empirical Study Leveraging Large Language Models

Small Language Model Can Self-correct

The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning

TurkishBERTweet: Fast and Reliable Large Language Model for Social Media Analysis

LoRA ensembles for large language model fine-tuning

Evaluating Instruction-Tuned Large Language Models on Code Comprehension and Generation

SPDF: Sparse Pre-training and Dense Fine-tuning for Large Language Models

Fundamentals of Generative Large Language Models and Perspectives in Cyber-Defense

Beyond Prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering Representations

Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning

Scaling Federated Learning for Fine-tuning of Large Language Models

SecureFixAgent: A Hybrid LLM Agent for Automated Python Static Vulnerability Repair

A Fast, Performant, Secure Distributed Training Framework For Large Language Model

Improving Influence-based Instruction Tuning Data Selection for Balanced Learning of Diverse Capabilities

RedditPersona: A Modular Framework for Community-Conditioned LLM Adaptation from Reddit

HermesHFL: Incentive-Compatible Hierarchical Federated Unlearning for Dynamic LLM Fine-Tuning

Benign Alone, Harmful Together: Exploiting Experience Composition in Self-Evolving LLM Agents

Device-First Feedback: Toward Mobile-Native LLM-Driven Neural Architecture Search

Strategy-Aware Parameter-Efficient Adaptation for LLM-based Auto-Bidding

MM-ARC: Multimodal Adaptive Routing of Capital with Robustness-Audited Strategy Pools

Endogenous Resistance to Activation Steering in Language Models