arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

AI 大模型

语言大模型 / LLM

大语言模型、预训练、指令微调、后训练和语言模型应用。

共收录 11714 信号源:cs.CL, cs.AI, cs.LG

1. 指令微调 11714 篇

2111.12386 2021-11-25 cs.CV 78%

One to Transfer All: A Universal Transfer Framework for Vision Foundation Model with Few Data

Yujie Wang, Junqin Huang, Mengya Gao, Yichao Wu, Zhenfei Yin, Ding Liang, Junjie Yan

专题命中 指令微调 :foundation model(title,abstract)

Comments Technical Report

详情

展开后加载摘要…

URL PDF HTML 收藏
2111.11904 2021-11-24 cs.SE 78%

Can Pre-trained Language Models be Used to Resolve Textual and Semantic Merge Conflicts?

Jialu Zhang, Todd Mytkowicz, Mike Kaufman, Ruzica Piskac, Shuvendu K. Lahiri

专题命中 指令微调 :language model(title,abstract)

详情

展开后加载摘要…

URL PDF HTML 收藏
2105.11108 2021-06-28 cs.IR 78%

Pre-trained Language Model based Ranking in Baidu Search

Lixin Zou, Shengqiang Zhang, Hengyi Cai, Dehong Ma, Suqi Cheng, Daiting Shi, Zhifan Zhu, Weiyue Su, Shuaiqiang Wang, Zhicong Cheng, Dawei Yin

专题命中 指令微调 :language model(title,abstract)

Comments 9-pages, 3 figures, 7 tables, SIGKDD 2021 accepted paper

详情

展开后加载摘要…

URL PDF HTML 收藏
2106.07920 2021-06-22 math.SG math.AG 78%

Lattice Formulas For Rational SFT Capacities

Julian Chaidez, Ben Wormleighton

专题命中 指令微调 :SFT(title,abstract)

Comments 35 pages, 11 figures, comments welcome! Made corrections to the statements of Lemma 10 and Theorem 5. Added more details to proofs in Section 4.5

详情

展开后加载摘要…

URL PDF HTML 收藏
2103.15505 2021-03-30 math.GR math.DS 78%

Veelike actions and the MCG of a mixing SFT

Ville Salo

专题命中 指令微调 :SFT(title,abstract)

Comments 5 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
1903.11884 2021-01-29 math.SG math.GT 78%

SFT computations and intersection theory in higher-dimensional contact manifolds

Agustin Moreno

专题命中 指令微调 :SFT(title,abstract)

Comments 50 pages. Streamlined from the author's PhD thesis

Journal ref Journal of the Institute of Mathematics of Jussieu, 25 January 2021

详情

展开后加载摘要…

URL PDF HTML 收藏
2008.04057 2020-09-21 cs.AI cs.CL cs.GT cs.LG 78%

The Chess Transformer: Mastering Play using Generative Language Models

David Noever, Matt Ciolino, Josh Kalin

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

Comments 7 Pages, 6 Figures, AAAI Format, AAAI 21

详情

展开后加载摘要…

URL PDF HTML 收藏
1806.07025 2018-08-16 math.SG 78%

Polyfold and SFT Notes I: A Primer on Polyfolds and Construction Tools

Joel W. Fish, Helmut Hofer

专题命中 指令微调 :SFT(title,abstract)

详情

展开后加载摘要…

URL PDF HTML 收藏
1711.11407 2017-12-01 eess.SP 78%

FPS-SFT: A Multi-dimensional Sparse Fourier Transform Based on the Fourier Projection-slice Theorem

Shaogang Wang, Vishal M. Patel, Athina Petropulu

专题命中 指令微调 :SFT(title,abstract)

详情

展开后加载摘要…

URL PDF HTML 收藏
1704.03633 2017-06-28 hep-th 78%

From the octagon to the SFT vertex - gluing and multiple wrapping

Zoltan Bajnok, Romuald A. Janik

专题命中 指令微调 :SFT(title,abstract)

Comments 25 pages, many small figures

详情

展开后加载摘要…

URL PDF HTML 收藏
1106.3914 2015-05-28 hep-th 78%

Analytic solutions for Dp branes in SFT

L. Bonora, S. Giaccari, D. D. Tolla

专题命中 指令微调 :SFT(title,abstract)

Comments 14 pages

Journal ref JHEP12(2011)033

详情

展开后加载摘要…

URL PDF HTML 收藏
1412.3466 2014-12-12 hep-th 78%

Renormalization schemes for SFT solutions

Joanna L. Karczmarek, Matheson Longton

专题命中 指令微调 :SFT(title,abstract)

Comments 41 pages, 1 figure

详情

展开后加载摘要…

URL PDF HTML 收藏
1412.0936 2014-12-03 hep-th 78%

Comments on lump solutions in SFT

Loriano Bonora, Driba D. Tolla

专题命中 指令微调 :SFT(title,abstract)

Comments 30 pages, no figures

详情

展开后加载摘要…

URL PDF HTML 收藏
0912.5457 2014-11-20 hep-th astro-ph.CO gr-qc 78%

SFT non-locality in cosmology: solutions, perturbations and observational evidences

Alexey S. Koshelev

专题命中 指令微调 :SFT(title,abstract)

Comments To be published in Proceedings of Invisible Universe 2009

Journal ref AIP Conf.Proc.1241:630-638,2010

详情

展开后加载摘要…

URL PDF HTML 收藏
1011.5672 2013-12-03 hep-th astro-ph.CO gr-qc 78%

Perturbative stability of SFT-based cosmological models

Federico Galli, Alexey S. Koshelev

专题命中 指令微调 :SFT(title,abstract)

Comments Version accepted for publicatin in JCAP, 19 pages, 6 figures, uses jcappub.sty

详情

展开后加载摘要…

URL PDF HTML 收藏
1109.4336 2012-05-21 hep-th 78%

Lump solutions in SFT. Complements

L. Bonora, S. Giaccari, D. D. Tolla

专题命中 指令微调 :SFT(title,abstract)

Comments 38 pages, expanded version, Table 3 corrected, App.B suppressed, sec.8 added

详情

展开后加载摘要…

URL PDF HTML 收藏
1109.6054 2011-09-29 math.AC 78%

Amalgamated algebra extensions defined by Von Neumann regular and SFT conditions

Khalid Louartiti, Najib Mahdou

专题命中 指令微调 :SFT(title,abstract)

详情

展开后加载摘要…

URL PDF HTML 收藏
1010.1773 2011-01-24 hep-th astro-ph.CO gr-qc 78%

Multi-scalar field cosmology from SFT: an exactly solvable approximation

Federico Galli, Alexey S. Koshelev

专题命中 指令微调 :SFT(title,abstract)

Comments Extended version of the proceedings of the Bogolyubov-2009 conference

Journal ref Theor.Math.Phys.164:1169-1175(2010); Teor.Mat.Fiz.164:401-409,2010

详情

展开后加载摘要…

URL PDF HTML 收藏
0902.4317 2009-12-01 math.SG 78%

Rational SFT, linearized Legendrian contact homology, and Lagrangian Floer cohomology

Tobias Ekholm

专题命中 指令微调 :SFT(title,abstract)

Comments 32 pages, 6 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2504.05995 2026-04-08 cs.CL cs.AI 77%

NativQA Framework: Enabling LLMs and VLMs with Native, Local, and Everyday Knowledge

NativQA框架:使LLMs和VLMs具备原生、本地和日常知识

Firoj Alam, Md Arid Hasan, Sahinur Rahman Laskar, Mucahid Kutlu, Kareem Darwish, Shammur Absar Chowdhury

机构 * Qatar Computing Research Institute, Qatar(卡塔尔计算研究所) University of Toronto, Canada(多伦多大学) UPES, India(印度UPES大学) Qatar University, Qatar(卡塔尔大学)

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

AI总结 本文提出NativQA框架,通过整合多模态数据,构建本地化问答数据集,提升不同语言和文化背景下的模型性能。

Comments LLMs, Native, Multilingual, Language Diversity, Contextual Understanding, Minority Languages, Culturally Informed, Foundation Models, Large Language Models

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.20971 2025-02-13 cs.CV cs.AI cs.LG 77%

BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks

Yunhan Zhao, Xiang Zheng, Lin Luo, Yige Li, Xingjun Ma, Yu-Gang Jiang

机构 * Fudan University(复旦大学) City University of Hong Kong(香港城市大学) Singapore Management University(新加坡管理大学)

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

Journal ref ICLR 2025, BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks. In Proceedings of the International Conference on Learning Representations (ICLR), 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.27101 2026-08-28 cs.AI 新提交 77%

pro-team at LLMs4OL 2026 Tasks Flagship and Reuse: Retrieval-Augmented Generation and Vocabulary-Constrained Filtering for Ontology Learning

LLMs4OL 2026任务的团队:旗舰任务与复用:面向本体学习的检索增强生成及词汇约束过滤

Shivam Mishra, Dhannu Ram Meena, Muneendra Ojha, Krishna Pratap Singh, Kuldeep Singh

机构 * Indian Institute of Information Technology Allahabad(印度阿拉哈巴德信息技术学院)

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

AI总结 该团队针对LLMs4OL 2026挑战赛的两个本体学习任务,采用检索增强生成与词汇约束过滤方法,取得了特定指标结果,但存在未提取非分类关系的局限。

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.15763 2026-08-28 cs.CL 版本更新 77%

Training Agents to Evolve with Their Harness: TaoLive Digital Avatar Agent Technical Report

TaoLive数字虚拟人智能体技术报告:训练智能体随其Harness进化

TaoLive AIGC LLM Team, Yuhan Sun, Wenhao Lin, Yongdong Luo, Yibo Hu, Meiguang Jin, Junfeng Ma, Weihang Pan, Jiaxin Zhao, Zulong Chen

机构 * TaoLive(陶境科技)

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

AI总结 本研究针对直播电商数字虚拟人主播的实时需求,提出HAT方法,结合HSA的三阶段训练,使35B紧凑模型在低延迟下适配Harness变化,性能优于基础模型及通用大语言模型。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.07737 2026-08-26 cs.CV cs.AI 版本更新 77%

Seeing vs. Believing: Evaluating the Language Bias of Open-Source MLLMs in Counter-Intuitive Scenes

看见 vs. 相信:评估开源多模态大模型在反直觉场景中的语言偏见

Chen Ling, Tongwei Zhang, Hanqian Li, Nai Ding

机构 * Zhejiang University(浙江大学) Beijing University of Posts and Telecommunications(北京邮电大学) Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州))

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

AI总结 为评估多模态大模型处理反直觉动作场景的能力,提出CAIT基准(400个高保真合成场景),发现开源模型因语言先验而忽视视觉证据,性能接近随机水平,而链式思维推理虽提升准确率但导致过度思考拒绝视觉内容,通过微调和结构化提示可缓解此偏见。

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.21909 2026-08-25 cs.LG 新提交 77%

CD-LoRA: Consistency-Driven Low-Rank Adaptation for Multi-Task Fine-Tuning

CD-LoRA:面向多任务微调的一致性驱动低秩适配

Qian Zha, Jinda Liu, Yuan Wu, Yi Chang

机构 * School of Artificial Intelligence, Jilin University(吉林大学人工智能学院) International Center of Future Science, Jilin University(吉林大学未来科学国际中心)

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

AI总结 该研究针对多任务LoRA方法存在的训练-推理不一致问题,提出无路由的CD-LoRA,通过一致性驱动对齐机制提升多任务微调的稳定性与性能,优于现有多适配器基线。

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.21408 2026-08-25 cs.AI 新提交 77%

Hate Speech Classification In Roman Urdu: A Comparative Study On Parameter Efficient Fine-Tuning And Prompt Engineering

罗马乌尔都语仇恨言论分类:参数高效微调与提示工程的对比研究

Toneema Zubair

专题命中 指令微调 :LLM(summary_cn,abstract_cn);分类 cs.AI

AI总结 本研究针对低资源的罗马乌尔都语,对比了LLM零样本推理、LoRA的PEFT、混合/人工提示调优、零/少样本提示工程四种仇恨言论分类方法,以确定最有效的技术。

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.14642 2026-08-25 cs.CL 77%

Speak-to-Structure: Evaluating LLMs in Open-domain Natural Language-Driven Molecule Generation

Speak-to-Structure:评估大语言模型在开放域自然语言驱动的分子生成中的表现

Jiatong Li, Junxian Li, Weida Wang, Yunqing Liu, Changmeng Zheng, Yatao Bian, Dongzhan Zhou, Xiao-yong Wei, Qing Li

机构 * Hong Kong Polytechnic University(香港理工大学) Shanghai Jiao Tong University(上海交通大学) Shanghai AI Lab(上海人工智能实验室) National University of Singapore(新加坡国立大学)

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

AI总结 提出Speak-to-Structure基准,通过分子编辑、优化和定制生成任务评估大语言模型在开放域自然语言驱动分子生成中的创造性能力,并引入OpenMolIns指令微调数据集使Llama3.1-8B超越GPT-4o等模型。

Comments Accepted by KDD 2026. Our codes and datasets are fully accessible through the https://github.com/phenixace/S2-TOMG-Bench and https://huggingface.co/datasets/phenixace/S2-TOMG-Bench

Journal ref Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '26), 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.03318 2026-08-25 cs.CL 版本更新 77%

MIRROR: A Multi-Agent Framework with Iterative Adaptive Revision and Hierarchical Retrieval for Optimization Modeling in Operations Research

MIRROR: 一种用于运筹学优化建模的具有迭代自适应修正与分层检索的多智能体框架

Yifan Shi, Jiayi Wang, Minyi Wu, Ye Fan, Jialong Shi, Jianyong Sun

机构 * Xi’an Jiaotong University(西安交通大学) Northwestern Polytechnical University(西北工业大学)

专题命中 指令微调 :large language model(abstract);language model(abstract);post-training(abstract);分类 cs.CL

AI总结 提出一种免微调的多智能体框架MIRROR,通过执行驱动的迭代自适应修正和分层检索机制,将自然语言优化问题直接转化为数学模型和求解器代码,在标准运筹学基准上优于现有方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.20830 2026-08-24 cs.CY cs.LG 新提交 77%

Fine-tuning LLMs for Tourist Trajectory Prediction using Field Experiment Data

利用实地实验数据微调大型语言模型(LLMs)进行游客轨迹预测

Tatsuya Amano, Hirozumi Yamaguchi

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

AI总结 该研究利用日本和歌山城公园的566条轨迹微调Llama-3.1-8B,实现49.1%的下一个兴趣点准确率,在样本不足场景泛化性强,为旅游轨迹预测提供了高保真行为模型及反事实分析基础。

Comments 5 pages, 4 figures. Accepted at the 2nd Workshop on AI for Urban Planning (AI4UP) at AAAI-26, Singapore, January 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.13069 2026-08-14 cs.AI 新提交 77%

Behavioral Reprogramming of Open-Weights Models: Cognitive Plasticity and Alignment Bounds

开放权重模型的行为重编程:认知可塑性与对齐边界

Lucia Malíčková

机构 * National Supercomputing Centre, Slovakia(斯洛伐克国家超级计算中心)

专题命中 指令微调 :large language model(abstract);language model(abstract);preference optimization(abstract);分类 cs.AI

AI总结 该研究通过大规模并行超参数搜索等方法,对开放权重LLMs进行行为重编程,实现了主动苏格拉底式对话框架,明确了PEFT的边界等关键结论,为跨语言行为修改提供了实证框架。

Comments Preprint submitted to arXiv, August 12, 2026. 13 pages, 5 figures

详情

展开后加载摘要…

URL PDF HTML 收藏