phic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Avoiding unsafe sets when training with Langevin Dynamics

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Adam Oberman [ view email ] [v1] Wed, 8 Jul 2026 15:36:45 UTC (184 KB) Full-text links: Access Paper: View a PDF of the paper titled Avoiding unsafe sets when training with Langevin Dynamics, by Adam M. Oberman View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs math stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers?

) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Fast Rates for Semi-Supervised Learning via Data-Augmentation Graph Regularization

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Adam Oberman [ view email ] [v1] Wed, 8 Jul 2026 15:11:21 UTC (76 KB) Full-text links: Access Paper: View a PDF of the paper titled Fast Rates for Semi-Supervised Learning via Data-Augmentation Graph Regularization, by Adam M. Oberman View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs math stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Tog

Telegram必备的搜索引擎，极搜JISOU帮你精准找到，想要的群组、频道、视频、音乐

👉 t.me/jisou2?start=a_8247614025

gle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

The Optimal Sample Complexity of Learning Autoregressive Chain-of

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Zhiyuan Li [ view email ] [v1] Wed, 8 Jul 2026 13:49:54 UTC (33 KB) Full-text links: Access Paper: View a PDF of the paper titled The Optimal Sample Complexity of Learning Autoregressive Chain-of-Thought, by Zhiyuan Li View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connect

ed Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Gauge-Invariant Learnable Spectral Positional Encodings for Directed Graphs via Hermitian Block Krylov Subspaces

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Jiaqing Xie [ view email ] [v1] Wed, 8 Jul 2026 06:05:54 UTC (57 KB) Full-text links: Access Paper: View a PDF of the paper titled Gauge-Invariant Learnable Spectral Positional Encodings for Directed Graphs via Hermitian Block Krylov Subspaces, by Jiaqing Xie and 1 other authors View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle

Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Local large deviations for linear-region growth in random piecewise

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Christian Hirsch [ view email ] [v1] Wed, 8 Jul 2026 05:14:19 UTC (150 KB) Full-text links: Access Paper: View a PDF of the paper titled Local large deviations for linear-region growth in random piecewise-linear networks, by Recep \"Ozkan and Christian Hirsch View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: math stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connecte

d Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

迁移学习与线性判别分析

近日，一篇题为“Transfer Learning for Linear Discriminant Analysis with a Shared Classification Signal”的研究论文在arXiv平台上提交。该论文由Yonghan Zhang等三位作者完成，主要聚焦于迁移学习在共享分类信号场景下与线性判别分析（LDA）的结合。研究旨在利用源域中共享的分类信号来提升目标域的分类性能，为统计学习中的迁移应用提供了新的理论视角和方法框架。论文目前以PDF格式公开，并提供了HTML预览及TeX源码，供学界深入探讨。该工作预计将对机器学习、统计分类以及跨领域知识迁移等领域产生积极影响。 #迁移学习 #线性判别分析 #共享分类信号 #arXiv #机器学习 #统计学 #论文

生成式代理助力最佳臂识别研究

一篇题为《Best-Arm Identification with Generative Proxy》的学术论文近日在arXiv预印本平台上发布。该论文由Tianyi Ma等四位作者共同撰写，提交于2026年7月8日。研究聚焦于强化学习中的最佳臂识别问题，提出了一种利用生成式代理（Generative Proxy）来提升识别效率的新方法。通过引入生成模型作为代理，该方法有望在更少的采样次数内准确识别出最优选项，从而降低实验成本并加速决策过程。这一创新为多臂老虎机及在线学习领域提供了新的理论视角和实用工具。 #机器学习 #最佳臂识别 #生成模型 #强化学习 #arXiv #学术论文 #AI

从跳跃到签名

据arXiv预印本显示，研究人员Niels Cariou-Kotlarek和Vasileios Lampos提出了一种名为“从跳跃到签名”的生成方法，用于时间点过程建模。该工作利用签名（signatures）技术对事件序列的跳跃特征进行表征，能够有效生成符合真实动态的时间点过程数据。这一方法有望在金融、流行病学、网络分析等领域的事件序列生成与预测中发挥重要作用。论文已提交至arXiv平台，目前可供查阅。 #时间点过程 #生成模型 #签名 #机器学习 #统计建模 #arXiv #学术研究 #事件序列

虚拟筛选校准中的静默失败：边际一致性预测遗漏少数类，条件性修复恢复之

一项新研究揭示了在虚拟筛选中使用边际一致性预测时存在的隐蔽缺陷。该方法在整体校准表现看似良好时，却会系统性遗漏少数类（如活性化合物），导致假阴性率升高。研究者提出了一种基于类别条件的修复策略，通过为每个类别分别进行校准，显著恢复了模型对少数类的覆盖能力。实验表明，该修复方法在保持全局校准性的同时，有效提升了稀有类别的预测可靠性，对药物发现等应用场景具有重要实践意义。 #机器学习 #虚拟筛选 #一致性预测 #类别不平衡 #药物发现 #AI #校准 #少数类

研究人员提出统一框架检测AI生成内容与伪影

近日，一篇发表于arXiv预印本平台的论文提出了一种名为“统一检测框架”的新方法，旨在识别AI相关内容和合成伪影。该框架由张西峰等三位研究人员共同开发，能够高效检测各类AI生成文本、图像及代码中的伪造痕迹。研究团队表示，当前AI内容泛滥，现有检测工具往往局限于特定类型，而该框架实现了跨模态的统一检测，有望提升虚假信息识别能力。论文已通过DataCite分配DOI，并提供了HTML、TeX源码及多种引用格式。该工作属于计算机科学与统计学交叉领域，为AI安全治理提供了新思路。 #AI检测 #统一框架 #伪影识别 #arXiv #人工智能安全 #计算机科学 #学术研究

统计逆学习与L1正则化研究登arXiv

近日，一篇题为《Statistical inverse learning and $\ell^1$-regularization》的学术论文在arXiv预印本平台发布。该论文由Abhishake Rastogi等人撰写，聚焦于统计逆学习问题中L1正则化方法的理论分析。研究涉及逆问题在统计学和机器学习中的应用，探讨了L1正则化在稀疏解恢复、噪声控制等方面的性质。论文提供了完整的数学推导和理论保证，对理解高维数据下的逆问题求解具有潜在价值。目前该文可通过arXiv免费获取，相关代码和数据链接也已开放。 #学术论文 #arXiv #统计逆学习 #L1正则化 #机器学习 #数学

DiPhon

一篇题为《DiPhon: Diffusion on Graphons for Scalable Graph Generation》的学术论文近日在arXiv预印本平台发布。该研究由Sergio Rozada等五位作者共同完成，提出了一种创新的图生成框架，将扩散模型与图on（Graphon）理论相结合，旨在解决现有图生成方法在大规模图数据上的可扩展性问题。图on作为无限图的极限对象，能够表示任意大小的图结构，而扩散模型在其中运行可有效捕捉图的全局与局部特征。该方法在保证生成质量的同时显著降低计算复杂度，为社交网络、生物信息学等领域的大规模图生成任务提供了新思路。论文已提供PDF全文及HTML实验版本，并附带代码链接，目前处于2026年7月的最新提交版本。 #图生成 #扩散模型 #图on #可扩展性 #arXiv #AI #论文 #图神经网络

面向隐马尔可夫和因子隐马尔可夫模型的张量化算法与可扩展滤波方法

近日，一篇来自arXiv的学术论文提出了一系列张量化的算法与可扩展的滤波方法，旨在解决隐马尔可夫模型（HMM）及因子隐马尔可夫模型（FHMM）在大规模数据下的计算效率问题。论文作者Roxana Barrios等人通过引入张量结构，对传统推理流程进行重构，显著降低了状态空间爆炸带来的计算复杂度。该方法不仅适用于经典HMM，还能高效处理FHMM中多链交互的滤波难题。实验表明，该框架在保持精度的前提下大幅提升了可扩展性，有望推动HMM在语音识别、生物信息学、时间序列分析等领域的实际应用。 #隐马尔可夫模型 #张量化 #滤波方法 #机器学习 #AI #学术论文 #算法 #可扩展性

随机凸优化问题中稳定点寻找方法研究论文发布

据arXiv预印本平台显示，一篇题为《Finding a stationary point of a stochastic convex problem》的学术论文于2026年7月8日提交。该论文由Felipe Areces、John Duchi和Malo Sommers共同撰写，聚焦于随机凸优化问题中稳定点的求解方法。研究有望为机器学习、统计学习等领域中的优化问题提供理论支撑，相关算法和分析对理解随机凸问题的收敛性具有重要意义。目前论文提供PDF和HTML版本，并附有引用工具及代码链接，供学术界参考。 #arXiv #机器学习 #凸优化 #随机算法 #论文 #优化理论 #学术研究