语言大模型 / LLM - arXivDaily 专题

2506.02295 2025-06-04 cs.CV cs.AI 86%

QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation

Ahmed Wasfy, Omer Nacar, Abdelakreem Elkhateb, Mahmoud Reda, Omar Elshehy, Adel Ammar, Wadii Boulila

机构 * NAMAA ； KAND CA Corp. ； Prince Sultan University

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

2505.12545 2025-05-22 cs.CL 86%

Towards Reliable and Interpretable Traffic Crash Pattern Prediction and Safety Interventions Using Customized Large Language Models

Yang Zhao, Pu Wang, Yibo Zhao, Hongru Du, Hao Frank Yang

机构 * Center for Systems Science and Engineering（系统科学与工程中心）； Johns Hopkins University（约翰霍普金斯大学）

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

Comments Last revised 13 Feb 2025. Under review in Nature portfolio

2503.12293 2025-05-16 cs.SE cs.LG 86%

Unified Modeling Language Code Generation from Diagram Images Using Multimodal Large Language Models

Averi Bates, Ryan Vavricka, Shane Carleton, Ruosi Shao, Chongle Pan

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

Comments Published in the Journal of Machine Learning with Applications, Author Contributions: Averi Bates: Methodology, Development, Analysis, Data Curation, Drafting, Review. Ryan Vavricka: Data Curation, Development, Review. Shane Carleton: Supervision, Funding. Ruosi Shao: Review. Chongle Pan: Supervision, Review

Journal ref Mach. Learn. Appl. 20 (2025) 100660

2411.14720 2025-04-03 cs.CL 86%

Optimizing Social Media Annotation of HPV Vaccine Skepticism and Misinformation Using Large Language Models: An Experimental Evaluation of In-Context Learning and Fine-Tuning Stance Detection Across Multiple Models

Luhang Sun, Varsha Pendyala, Yun-Shiuan Chuang, Shanglin Yang, Jonathan Feldman, Andrew Zhao, Munmun De Choudhury, Sijia Yang, Dhavan Shah

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

2502.20897 2025-03-03 cs.CL 86%

Beyond Demographics: Fine-tuning Large Language Models to Predict Individuals' Subjective Text Perceptions

Matthias Orlikowski, Jiaxin Pei, Paul Röttger, Philipp Cimiano, David Jurgens, Dirk Hovy

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

Comments Reviewed ARR December 2024

2502.08008 2025-02-17 cs.LG cs.CR 86%

An Interactive Framework for Implementing Privacy-Preserving Federated Learning: Experiments on Large Language Models

Kasra Ahmadi, Rouzbeh Behnia, Reza Ebrahimi, Mehran Mozaffari Kermani, Jeremiah Birrell, Jason Pacheco, Attila A Yavuz

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

2502.06633 2025-02-11 cs.SE cs.AI 86%

Combining Large Language Models with Static Analyzers for Code Review Generation

Imen Jaoua, Oussama Ben Sghaier, Houari Sahraoui

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

2412.18260 2025-01-07 cs.CL 86%

Investigating Large Language Models for Code Vulnerability Detection: An Experimental Study

Xuefeng Jiang, Lvhua Wu, Sheng Sun, Jia Li, Jingjing Xue, Yuwei Wang, Tingting Wu, Min Liu

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

Comments Under Review

2412.18419 2024-12-25 cs.AI 86%

Research on the Proximity Relationships of Psychosomatic Disease Knowledge Graph Modules Extracted by Large Language Models

Zihan Zhou, Ziyi Zeng, Wenhao Jiang, Yihui Zhu, Jiaxin Mao, Yonggui Yuan, Min Xia, Shubin Zhao, Mengyu Yao, Yunqian Chen

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

2305.13862 2024-11-12 cs.CL 86%

A Trip Towards Fairness: Bias and De-Biasing in Large Language Models

Leonardo Ranaldi, Elena Sofia Ruzzetti, Davide Venditti, Dario Onorati, Fabio Massimo Zanzotto

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

Journal ref 2024.starsem-1.30

2410.18035 2024-10-24 cs.CL 86%

MiLoRA: Efficient Mixture of Low-Rank Adaptation for Large Language Models Fine-tuning

Jingfan Zhang, Yi Zhao, Dan Chen, Xing Tian, Huanran Zheng, Wei Zhu

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

Comments Accepted by EMNLP 2024 Findings. arXiv admin note: substantial text overlap with arXiv:2405.18203

2407.14609 2024-07-23 cs.CL 86%

Adversarial Databases Improve Success in Retrieval-based Large Language Models

Sean Wu, Michael Koo, Li Yo Kao, Andy Black, Lesley Blum, Fabien Scalzo, Ira Kurtz

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

Comments 24 pages, 3 figures, 11 tables

2403.07794 2024-07-04 cs.CL 86%

Fine-tuning Large Language Models with Sequential Instructions

Hanxu Hu, Simon Yu, Pinzhen Chen, Edoardo M. Ponti

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

Comments 21pages, 8 figures

2406.04879 2024-06-10 cs.CL 86%

A Deep Dive into the Trade-Offs of Parameter-Efficient Preference Alignment Techniques

Megh Thakkar, Quentin Fournier, Matthew D Riemer, Pin-Yu Chen, Amal Zouaq, Payel Das, Sarath Chandar

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

Comments Accepted to ACL (Main) 2024

2403.03129 2024-06-07 cs.CL 86%

CoGenesis: A Framework Collaborating Large and Small Language Models for Secure Context-Aware Instruction Following

Kaiyan Zhang, Jianyu Wang, Ermo Hua, Biqing Qi, Ning Ding, Bowen Zhou

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

Comments Accepted to ACL 2024 (Main Conference)

2403.01754 2024-03-05 cs.CL 86%

Derivative-Free Optimization for Low-Rank Adaptation in Large Language Models

Feihu Jin, Yin Liu, Ying Tan

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

Comments 14 pages, 4 figures, 5 tables

2402.06053 2024-02-12 cs.HC cs.AI cs.CY 86%

Randomness Is All You Need: Semantic Traversal of Problem-Solution Spaces with Large Language Models

Thomas Sandholm, Sayandev Mukherjee, Bernardo A. Huberman

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

2309.07900 2024-01-31 cs.CL cs.IR 86%

Ambiguity-Aware In-Context Learning with Large Language Models

Lingyu Gao, Aditi Chaudhary, Krishna Srinivasan, Kazuma Hashimoto, Karthik Raman, Michael Bendersky

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

Comments 15 pages in total

2312.15603 2023-12-27 cs.CL 86%

A Split-and-Privatize Framework for Large Language Model Fine-Tuning

Xicong Shen, Yang Liu, Huiqi Liu, Jue Hong, Bing Duan, Zirui Huang, Yunlong Mao, Ye Wu, Di Wu

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

2306.10165 2023-06-21 cs.CL 86%

Data Selection for Fine-tuning Large Language Models Using Transferred Shapley Values

Stephanie Schoch, Ritwick Mishra, Yangfeng Ji

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

Comments Accepted to ACL SRW 2023

2608.17719 2026-08-19 cs.SE cs.AI cs.CL 新提交 86%

What Aggregate Scores Miss: Measuring Item-Level Regressions in Commercial LLM API Migrations

聚合分数所遗漏的：商业大语言模型API迁移中项目级回归的测量

Xiaonan Xu, Wenjing Wu

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

AI总结该研究针对GPT-5.4到GPT-5.6 Sol的三次升级，发现聚合分数遗漏了项目级双向变化，发布了响应级档案与项目级评分输出。

Comments 25 pages, 1 figure, 10 tables (including 8 appendix tables)

2608.14693 2026-08-18 cs.CL cs.AI 新提交 86%

Domain Agnostic Text Redaction from Natural Language Rules using Instruction Tuning

基于指令微调的自然语言规则的领域无关文本脱敏

Aravindhan Arunagiri, Ayaan Khan, Udayaadithya Avadhanam, SaiBarath Sundar

机构 * Mphasis Limited（美费西斯有限公司）

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

AI总结本文提出一种基于指令微调语言模型的领域无关可解释文本脱敏方案，支持用户以自然语言定义敏感信息，可应用于法律、医疗等关键场景的自动文本脱敏。

2607.27940 2026-08-03 cs.LG cs.CL 版本更新 86%

TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement

TriShield：通过正交梯度投影与优化器状态纠缠实现联邦语言模型微调中隐私后门的零效用损失防御

Cheng Wei

机构 * Honor Device Co., Ltd.（荣耀终端有限公司）

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

AI总结 TriShield是一种三层确定性防御，通过参数特征检测、有状态虚拟迭代和零效用正交投影，可完全抵御NeuroImprint攻击，且零模型效用损失、无额外通信轮次，计算开销低。

Comments 12 pages,3 figures

2607.23322 2026-07-28 cs.CL cs.CY cs.ET cs.LG 新提交 86%

IKS-Instruct: A 24,000-Example Multilingual Dataset for Teaching Language Models Indian Knowledge Systems

IKS-Instruct：一个用于教授语言模型印度知识体系的包含24000个示例的多语言数据集

Shwetha Singaravelu, Gayathri Muruganantham, Lakshmi Rajendran, Santhosh Sivasubramani

机构 * Intrinsic Lab, Centre for Sensors, Instrumentation and Cyber-Physical System Engineering (Centre for SeNSE), Indian Institute of Technology Delhi（印度理工学院德里分校传感器、仪器仪表和网络物理系统工程中心（传感中心）本征实验室）； RSL Quantum, FITT, IIT Delhi（印度理工学院德里分校 FITT 学院 RSL 量子实验室）

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

AI总结研究针对现有指令数据集问题，提出IKS-Instruct多语言数据集教授语言模型印度知识体系，源于六种源类型，经多评判评估框架评估质量，展示了该数据集在特定模型微调上的效果及模型质量与数据整理的关系。

Comments 32 pages, 5 figures

2603.03536 2026-07-28 cs.CL cs.AI cs.IR 版本更新 86%

SafeCRS: Personalized Safety Alignment for LLM-Based Conversational Recommender Systems

SafeCRS: 为基于大语言模型的对话推荐系统实现个性化安全对齐

Haochang Hao, Yifan Xu, Xinzhuo Li, Yingqiang Ge, Lu Cheng

机构 * University of Illinois at Chicago（伊利诺伊大学芝加哥分校）； University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）； Amazon（亚马逊）

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

AI总结 SafeCRS通过整合安全监督微调与安全组奖励解耦归一化策略优化，提升基于大语言模型的对话推荐系统在个性化安全约束下的推荐质量与安全对齐能力。

Comments Accepted by SIGKDD 2026

2510.07074 2026-07-28 cs.CL cs.AI 版本更新 86%

LuxInstruct: A Cross-Lingual Instruction Tuning Dataset For Luxembourgish

LuxInstruct：一个用于卢森堡语的跨语言指令微调数据集

Fred Philippy, Laura Bernardy, Siwen Guo, Jacques Klein, Tegawendé F. Bissyandé

机构 * SnT, University of Luxembourg（卢森堡大学SnT学院）； Zortify Labs, Zortify S.A.（Zortify实验室）

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

AI总结针对卢森堡语缺乏高质量指令数据集问题，利用英、法、德对齐数据创建跨语言指令微调数据集，避免机器翻译弊端，提升模型在卢森堡语中的生成能力，为低资源语言发展提供新思路。

Comments Paper accepted at KONVENS 2026

2607.19674 2026-07-23 cs.CR cs.AI cs.LG 新提交 86%

FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defense

FedLSG：用于联邦图后门防御的大语言模型增强语义校准

Chenyu Zhou, Yabin Peng, Wei Huang, Kunlin Li, Shuaishuai Zhang, Xinyuan Miao

机构 * Southeast University（东南大学）； Purple Mountain Laboratories（紫金山实验室）； Institute of AI for Industries（人工智能产业研究院）； Chinese Academy of Sciences（中国科学院）

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

AI总结针对联邦图神经网络易受后门攻击问题，提出FedLSG框架，将大语言模型集成防御，通过图与行为到文本的转换及轻量级师生架构，在不损图完整性时显著提升对后门攻击的抵抗力。

2607.07916 2026-07-10 cs.AI cs.LG 新提交 86%

Persona Cartography: Charting Language Model Personality Traits in Weight Space

人物角色制图：在权重空间中描绘语言模型的个性特征

Luke Baines, Anton Gonzalvez Hawthorne, Mariia Koroliuk, Irakli Shalibashvili, Clément Dumas, Konstantinos Voudouris, David Demitri Africa

机构 * dsit.gov.uk（数字科学与创新部）

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

AI总结研究大型语言模型人物角色，用大五人格框架将其视为行为特征空间位置，训练低秩适配器控制特征，评估其效果，发现可构建混合人物角色并保持性能，还表明特征轴影响安全行为，引入无监督管道，为人格测量等架桥。

Comments 85 pages, 80 figures

2606.31166 2026-07-01 cs.CL cs.LG 新提交 86%

TAG-DLM: Diffusion Language Models for Text-Attributed Graph Learning

TAG-DLM：面向文本属性图学习的扩散语言模型

Lingjie Chen, Yuanchen Bei, Haobo Xu, Yanjun Zhao, Yuzhong Chen, Hanghang Tong

机构 * UIUC（伊利诺伊大学厄巴纳-香槟分校）； VISA（VISA研究院）

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

AI总结提出TAG-DLM方法，通过掩码扩散语言模型统一文本推理与图消息传递，线性化邻域并利用拓扑注意力掩码实现图结构注入，无需微调即可适应节点分类、链接预测等任务，性能优于现有基线。

2606.30266 2026-06-30 cs.LG cs.AI 86%

Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation

迈向连续运动-语言智能体：用于增量运动理解与生成的LoRA变体

Bertram Taetz, Hugo Albuquerque Cosme da Silva, Gabriele Bleser-Taetz

机构 * IT & Engineering International University of Applied Sciences（IT与工程国际应用科学大学）

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

AI总结针对运动-语言智能体在连续任务中学习新运动概念时灾难性遗忘的问题，提出基于LoRA的混合专家架构，通过自编码器路由器选择任务特定专家，无需任务标签，在HumanML3D基准上实现双向任务近零遗忘并保持高质量。

Comments 16 pages, 1 figure, Accepted at the Conference on Lifelong Learning Agents (CoLLAs) 2026

1. 指令微调 11585 篇

QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation

Towards Reliable and Interpretable Traffic Crash Pattern Prediction and Safety Interventions Using Customized Large Language Models

Unified Modeling Language Code Generation from Diagram Images Using Multimodal Large Language Models

Optimizing Social Media Annotation of HPV Vaccine Skepticism and Misinformation Using Large Language Models: An Experimental Evaluation of In-Context Learning and Fine-Tuning Stance Detection Across Multiple Models

Beyond Demographics: Fine-tuning Large Language Models to Predict Individuals' Subjective Text Perceptions

An Interactive Framework for Implementing Privacy-Preserving Federated Learning: Experiments on Large Language Models

Combining Large Language Models with Static Analyzers for Code Review Generation

Investigating Large Language Models for Code Vulnerability Detection: An Experimental Study

Research on the Proximity Relationships of Psychosomatic Disease Knowledge Graph Modules Extracted by Large Language Models

A Trip Towards Fairness: Bias and De-Biasing in Large Language Models

MiLoRA: Efficient Mixture of Low-Rank Adaptation for Large Language Models Fine-tuning

Adversarial Databases Improve Success in Retrieval-based Large Language Models

Fine-tuning Large Language Models with Sequential Instructions

A Deep Dive into the Trade-Offs of Parameter-Efficient Preference Alignment Techniques

CoGenesis: A Framework Collaborating Large and Small Language Models for Secure Context-Aware Instruction Following

Derivative-Free Optimization for Low-Rank Adaptation in Large Language Models

Randomness Is All You Need: Semantic Traversal of Problem-Solution Spaces with Large Language Models

Ambiguity-Aware In-Context Learning with Large Language Models

A Split-and-Privatize Framework for Large Language Model Fine-Tuning

Data Selection for Fine-tuning Large Language Models Using Transferred Shapley Values

What Aggregate Scores Miss: Measuring Item-Level Regressions in Commercial LLM API Migrations

Domain Agnostic Text Redaction from Natural Language Rules using Instruction Tuning

TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement

IKS-Instruct: A 24,000-Example Multilingual Dataset for Teaching Language Models Indian Knowledge Systems

SafeCRS: Personalized Safety Alignment for LLM-Based Conversational Recommender Systems

LuxInstruct: A Cross-Lingual Instruction Tuning Dataset For Luxembourgish

FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defense

Persona Cartography: Charting Language Model Personality Traits in Weight Space

TAG-DLM: Diffusion Language Models for Text-Attributed Graph Learning

Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation