语言大模型 / LLM - arXivDaily 专题

2407.12036 2024-11-13 cs.CL cs.CV 91%

Exploring Advanced Large Language Models with LLMsuite

Giorgio Roffo

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

Comments Keywords: Language Model Benchmarking, Pre-Trained LLM Comparison, LLM Performance Analysis, NLP Model Evaluation Tools, Public Dataset Inference for LLMs, BLEU and ROUGE Metrics for LLM, Open Source LLM Testing Tools, Large Language Model Evaluation Software, NLP Benchmarking Suite, Comprehensive LLM Evaluation Toolkit

2608.10276 2026-08-12 cs.HC cs.CY 新提交 91%

Fine-Tuning Large Language Models for Codebook-Guided Coding of Students' Mathematics Metaphor Responses

针对学生数学隐喻回答的密码本引导编码任务微调大型语言模型

Liang Zhang, Stephen Hwang, Yue Ma, Jinfa Cai

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

AI总结本研究通过LoRA微调开放权重LLM，使其在学生数学隐喻的密码本引导编码任务中性能提升，表现媲美甚至超越仅提示的专有模型，可用于数学教育的规模化AI辅助测量。

2606.05194 2026-07-10 cs.LG cs.AI cs.CL 版本更新 91%

Temporal Preference Concepts and their Functions in a Large Language Model

时间偏好概念及其在大语言模型中的功能

Ian Rios-Sialer, Shantanu Darveshi, Shuai Jiang, Avigya Paudel, Anastasiia Pronina, Ipshita Bandyopadhyay, Justin Shenk

机构 * AISC（AI Safety Camp）； SPAR（Supervised Program for Alignment Research）

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

AI总结通过因果定位和激活修补，本文发现大语言模型在中间到上层节点编码时间偏好几何结构，且行为分析表明模型对未来折扣比人类更平缓，但偏好不稳定，可通过引导向量调控。

2605.12968 2026-05-15 cs.LG cs.AI cs.CL cs.LO 91%

Controlling Logical Collapse in LLMs via Algebraic Ontology Projection over F2

通过代数本体投影控制LLM中的逻辑崩溃

Hisashi Miyashita, Mgnite Inc

机构 * Mgnite Inc（Mgnite公司）

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

AI总结本文提出代数本体投影(AOP)，通过在GF(2)下投影LLM隐藏状态，利用42对关系作为代数密钥，提升零样本包含准确性，揭示逻辑一致性退化问题。

2605.00421 2026-05-04 cs.CL cs.AI cs.LG 91%

RadLite: Multi-Task LoRA Fine-Tuning of Small Language Models for CPU-Deployable Radiology AI

RadLite：用于CPU部署的放射学AI的多任务LoRA微调小型语言模型

Pankaj Gupta, Kartik Bose

机构 * Postgraduate Institute of Medical Education and Research（医学教育与研究研究生院）

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

AI总结研究通过LoRA微调小型语言模型，实现放射学多任务性能，展示在消费级CPU上部署的可行性，提出RadLite模型。

2602.08377 2026-04-28 cs.LG cs.AI cs.CL 91%

Reinforcement Learning with Backtracking Feedback

具有回溯反馈的强化学习

Bilgehan Sel, Vaishakh Keshava, Phillip Wallis, Lukas Rutishauser, Ming Jin, Dingcheng Li

机构 * Google（谷歌）； Google DeepMind（谷歌DeepMind）； Virginia Tech（弗吉尼亚理工大学）

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

AI总结本文提出RLBF框架，通过强化学习阶段使模型动态纠正生成错误，结合改进的SFT数据生成策略，提升LLM在对抗攻击和内在错误中的安全性。

Comments NeurIPS 2025

2412.02904 2026-04-21 cs.CL cs.AI cs.LG 91%

Enhancing Trust in Large Language Models via Uncertainty-Calibrated Fine-Tuning

通过不确定性校准微调增强大语言模型的信任

Ranganath Krishnan, Piyush Khanna, Omesh Tickoo

机构 * Capital One, AI Labs（Capital One人工智能实验室）； Wayve Technologies（Wayve技术公司）； Intel Corporation（英特尔公司）

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

AI总结本文提出一种不确定性感知微调方法，用于提升大语言模型在自然语言生成任务中的不确定性估计能力，从而提高生成响应的可信度并减少幻觉现象。

Comments ICLR 2026 Trustworthy AI workshop

2510.17516 2026-04-14 cs.CL cs.AI cs.CY cs.LG 91%

SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviors

SimBench：大型语言模型模拟人类行为能力的基准测试

Tiancheng Hu, Joachim Baumann, Lorenzo Lupo, Nigel Collier, Dirk Hovy, Paul Röttger

机构 * University of Cambridge（剑桥大学）； Stanford University（斯坦福大学）； Bocconi University（博科尼大学）； University of Oxford（牛津大学）

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

AI总结 SimBench通过统一20个多样化的数据集，评估大型语言模型模拟人类行为的 fidelity，发现模型性能与模型大小呈对数线性关系，但与计算资源无关，且存在指令微调与模拟准确性之间的权衡。

Comments Accepted at ICLR 2026. Project Website: http://simbench.tiancheng.hu/ Data: https://huggingface.co/datasets/pitehu/SimBench

2506.07503 2026-03-11 cs.SE 91%

Evaluating Large Language Models for Multilingual Vulnerability Detection at Dual Granularities

评估大型语言模型在双粒度多语言漏洞检测中的性能

Honglin Shu, Michael Fu, Junji Yu, Dong Wang, Chakkrit Tantithamthavorn, Junjie Chen, Yasutaka Kamei

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

AI总结本研究评估了大型语言模型在多语言漏洞检测中的性能，发现GPT-4o在函数级和行级检测中表现优异，优于其他模型。

Comments 53 pages, 10 figures

2602.22124 2026-02-26 cs.SE cs.AI cs.CL cs.LG 91%

SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents

SWE-Protégé: 通过选择性协作专家解锁小型语言模型作为软件工程代理

Patrick Tser Jern Kon, Archana Pradeep, Ang Chen, Alexander P. Ellis, Warren Hunt, Zijian Wang, John Yang, Samuel Thompson

机构 * Meta ； University of Michigan（密歇根大学）； Stanford University（斯坦福大学）

专题命中指令微调：language model(title,abstract)；small language model(title,abstract)；SLM(abstract)；post-training(abstract)

AI总结 SWE-Protégé通过选择性协作专家，使小型语言模型在软件工程任务中取得显著提升，实现42.4%的Pass@1成绩。

2510.17402 2025-10-21 cs.CL cs.AI cs.LG 91%

Leveraging Group Relative Policy Optimization to Advance Large Language Models in Traditional Chinese Medicine

Jiacheng Xie, Shuai Zeng, Yang Yu, Xiaoting Tang, Guanghui An, Dong Xu

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

2510.09400 2025-10-13 cs.SE 91%

TIT: A Tree-Structured Instruction Tuning Approach for LLM-Based Code Translation

He Jiang, Yufu Wang, Hao Lin, Peiyu Zou, Zhide Zhou, Ang Jia, Xiaochen Li, Zhilei Ren

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

2510.01624 2025-10-03 cs.LG cs.AI cs.CL 91%

Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead

Feiyang Kang, Michael Kuchnik, Karthik Padthe, Marin Vlastelica, Ruoxi Jia, Carole-Jean Wu, Newsha Ardalani

机构 * FAIR at Meta（Meta 的 FAIR）； Meta

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

Comments Preprint. Under Review

2509.09689 2025-09-15 cs.IR cs.AI cs.CL cs.LG 91%

Personas within Parameters: Fine-Tuning Small Language Models with Low-Rank Adapters to Mimic User Behaviors

Himanshu Thakur, Eshani Agrawal, Smruthi Mukund

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

2508.16677 2025-08-26 cs.LG cs.AI cs.CL 91%

Recall-Extend Dynamics: Enhancing Small Language Models through Controlled Exploration and Refined Offline Integration

Zhong Guan, Likang Wu, Hongke Zhao, Jiahui Wang, Le Wu

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

2406.08391 2025-08-19 cs.LG cs.AI cs.CL stat.ML 91%

Large Language Models Must Be Taught to Know What They Don't Know

Sanyam Kapoor, Nate Gruver, Manley Roberts, Katherine Collins, Arka Pal, Umang Bhatt, Adrian Weller, Samuel Dooley, Micah Goldblum, Andrew Gordon Wilson

机构 * New York University（纽约大学）； Abacus AI ； Cambridge University（剑桥大学）； Columbia University（哥伦比亚大学）

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

Comments NeurIPS 2024 Camera Ready

2508.04846 2025-08-08 cs.AI cs.CL cs.LG 91%

Fine-Tuning Small Language Models (SLMs) for Autonomous Web-based Geographical Information Systems (AWebGIS)

Mahdi Nazari Ashani, Ali Asghar Alesheikh, Saba Kazemi, Kimya Kheirkhah, Yasin Mohammadi, Fatemeh Rezaie, Amir Mahdi Manafi, Hedieh Zarkesh

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

2507.06528 2025-07-10 cs.CL cs.AI cs.ET cs.LG 91%

InvestAlign: Overcoming Data Scarcity in Aligning Large Language Models with Investor Decision-Making Processes under Herd Behavior

Huisheng Wang, Zhuoshi Pan, Hangjing Zhang, Mingxiao Liu, Hanqing Gao, H. Vicky Zhao

机构 * Department of Automation, Tsinghua University（自动化系，清华大学）

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

2505.07500 2025-05-13 cs.CV 91%

Learning to Reason and Navigate: Parameter Efficient Action Planning with Large Language Models

Bahram Mohammadi, Ehsan Abbasnejad, Yuankai Qi, Qi Wu, Anton Van Den Hengel, Javen Qinfeng Shi

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

2410.21533 2025-03-18 cs.LG cs.AI cs.CL stat.ML 91%

L3Ms -- Lagrange Large Language Models

Guneet S. Dhillon, Xingjian Shi, Yee Whye Teh, Alex Smola

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

Comments International Conference on Learning Representations (ICLR), 2025

2503.01449 2025-03-04 cs.SE 91%

Benchmarking Large Language Models for Multi-Language Software Vulnerability Detection

Ting Zhang, Chengran Yang, Yindu Su, Martin Weyssow, Hung Nguyen, Tan Bui, Hong Jin Kang, Yikun Li, Eng Lieh Ouh, Lwin Khin Shar, David Lo

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

2502.05562 2025-02-11 cs.DB 91%

Can Large Language Models Be Query Optimizer for Relational Databases?

Jie Tan, Kangfei Zhao, Rui Li, Jeffrey Xu Yu, Chengzhi Piao, Hong Cheng, Helen Meng, Deli Zhao, Yu Rong

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

Comments 15 pages

2502.01187 2025-02-04 cs.AI cs.CL cs.LG 91%

Skewed Memorization in Large Language Models: Quantification and Decomposition

Hao Li, Di Huang, Ziyu Wang, Amir M. Rahmani

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

2406.11675 2025-01-28 cs.LG cs.AI cs.CL stat.ML 91%

BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models

Yibin Wang, Haizhou Shi, Ligong Han, Dimitris Metaxas, Hao Wang

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

Comments Accepted at NeurIPS 2024. Additional experiments have been included in the appendix

2403.07384 2024-12-06 cs.CL cs.AI cs.LG 91%

SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models

Yu Yang, Siddhartha Mishra, Jeffrey N Chiang, Baharan Mirzasoleiman

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

2411.05193 2024-11-28 cs.LG cs.AI cs.CL 91%

Q-SFT: Q-Learning for Language Models via Supervised Fine-Tuning

Joey Hong, Anca Dragan, Sergey Levine

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

Comments 17 pages, 4 figures

2411.15203 2024-11-26 cs.LG cs.AI cs.CL 91%

Multimodal large language model for wheat breeding: a new exploration of smart breeding

Guofeng Yang, Yu Li, Yong He, Zhenjiang Zhou, Lingzhen Ye, Hui Fang, Yiqi Luo, Xuping Feng

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

2411.06767 2024-11-12 cs.CL cs.AI cs.LG 91%

PDC & DM-SFT: A Road for LLM SQL Bug-Fix Enhancing

Yiwen Duan, Yonghong Yu, Xiaoming Zhao, Yichang Wu, Wenbo Liu

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

Comments COLING-Industry 2025 accepted

2410.15570 2024-10-22 cs.CL cs.AI cs.LG 91%

Stacking Small Language Models for Generalizability

Laurence Liang

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

2410.09181 2024-10-15 cs.CR cs.AI cs.CL cs.CY cs.LG 91%

Can a large language model be a gaslighter?

Wei Li, Luyao Zhu, Yang Song, Ruixi Lin, Rui Mao, Yang You

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

Comments 10/26 (Main Body/Total), 8 figures

1. 指令微调 11601 篇

Exploring Advanced Large Language Models with LLMsuite

Fine-Tuning Large Language Models for Codebook-Guided Coding of Students' Mathematics Metaphor Responses

Temporal Preference Concepts and their Functions in a Large Language Model

Controlling Logical Collapse in LLMs via Algebraic Ontology Projection over F2

RadLite: Multi-Task LoRA Fine-Tuning of Small Language Models for CPU-Deployable Radiology AI

Reinforcement Learning with Backtracking Feedback

Enhancing Trust in Large Language Models via Uncertainty-Calibrated Fine-Tuning

SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviors

Evaluating Large Language Models for Multilingual Vulnerability Detection at Dual Granularities

SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents

Leveraging Group Relative Policy Optimization to Advance Large Language Models in Traditional Chinese Medicine

TIT: A Tree-Structured Instruction Tuning Approach for LLM-Based Code Translation

Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead

Personas within Parameters: Fine-Tuning Small Language Models with Low-Rank Adapters to Mimic User Behaviors

Recall-Extend Dynamics: Enhancing Small Language Models through Controlled Exploration and Refined Offline Integration

Large Language Models Must Be Taught to Know What They Don't Know

Fine-Tuning Small Language Models (SLMs) for Autonomous Web-based Geographical Information Systems (AWebGIS)

InvestAlign: Overcoming Data Scarcity in Aligning Large Language Models with Investor Decision-Making Processes under Herd Behavior

Learning to Reason and Navigate: Parameter Efficient Action Planning with Large Language Models

L3Ms -- Lagrange Large Language Models

Benchmarking Large Language Models for Multi-Language Software Vulnerability Detection

Can Large Language Models Be Query Optimizer for Relational Databases?

Skewed Memorization in Large Language Models: Quantification and Decomposition

BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models

SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models

Q-SFT: Q-Learning for Language Models via Supervised Fine-Tuning

Multimodal large language model for wheat breeding: a new exploration of smart breeding

PDC & DM-SFT: A Road for LLM SQL Bug-Fix Enhancing

Stacking Small Language Models for Generalizability

Can a large language model be a gaslighter?