Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE

Focus to learn more arXiv-issued DOI via DataCite Submission history From: Haozhan Tang [ view email ] [v1] Wed, 8 Jul 2026 06:23:42 UTC (699 KB) Full-text links: Access Paper: View a PDF of the paper titled Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE, by Haozhan Tang and 4 other authors View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs 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 Pa

pers? ) 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 great AI data centre cover

Then $75 per month. Complete digital access to quality FT journalism on any device. Cancel anytime during your trial. Access to eight surprising articles a day, hand-picked by FT editors. For seamless reading, access content via the FT Edit page on and receive the FT Edit newsletter. Essential digital access to quality FT journalism on any device. Pay a year upfront and save 20%. Complete digital access to quality FT journalism with expert analysis from industry leaders. Pay a year upfront and save 20%. Check whether you already have access via your university or organisation. Terms & Conditions apply Discover all the plans currently available in your country Digital access for organisations. Includes exclusive features and content. See why over a million readers pay to read the Financial Times.

Mercor acquires Deeptune to build AI training environments

Share Today's frontier AI models can pass nearly any exam you give them, but put an agent inside a real workflow and it stalls on the first task an entry-level employee could handle. This is because the model lacks experience. Experience comes from doing real work: navigating messy documents, legacy enterprise software, and sprawling company context while making hundreds of small decisions along the way. It means knowing what to do when something breaks or when the next step isn't obvious. A model learns to perform well only by practicing repeatedly in the environment where the work takes place. Building environments rigorous enough for that kind of training is incredibly hard. That's where Deeptune excels. Reinforcement learning has reached the point where a model can learn almost any task that can be clea

rly defined and scored. The constraint has shifted to the environments themselves: the places where models practice the work and get measured on whether they did it well. Building environments like these is exceptionally difficult. Each one has to faithfully recreate the software people use, translate expert judgment into realistic tasks, and measure performance clearly enough for a model to improve. Mercor's experts are already doing much of this work by writing evals. Deeptune provides the missing piece: the software platform where that expertise becomes realistic training environments at scale. Every environment has three parts: the software where the work takes place, the tasks that define what the agent must accomplish, and the verifiers that determine whether it succeeded. Mercor has built the expert layer behind that system. Our network of more than five million domain experts creates the tasks and verifiers that translate human judgment into learning signals for frontier models. Through APEX, we measure model performance on economically valuable tasks across thousands of real-world workflows. Deeptune builds the software layer. Over the past two years, the team has recreated hundreds of enterprise applications — from spreadsheets to Salesforce — and built some of the most sophisticated training environments used by frontier AI labs. We've seen the quality of their work firsthand because Mercor is already a Deeptune customer. Together, Mercor's expert network and Deeptune's software platform make it possible to build realistic training environments across far more industries, roles, and workflows than either company could alone. None of that happens without the people building it. I first met Tim, founder and CEO of Deeptune, two years ago, long before we started talking about working together. I was impressed by the technical ambition of Deeptune and Tim's conviction that environments would become the defining bottleneck for AI, long before most of the industry saw it. He and his team executed on that vision with remarkable speed. That's why I became an early investor, and ultimately why it felt obvious to bring our companies together. Deeptune is a small, focused team of engineers and operators with deep ties to frontier research. Earlier this year, Andreessen Horowitz led Deeptune's $43 million Series A. Tim and the entire Deeptune team are joining Mercor as we expand our team in NYC. For the labs we work with, this means environments that ship faster and train better. For enterprises deploying agents into their own workflows, it means one partner that can build the environment, staff the experts, and tell you whether the agent is ready. Every frontier lab is scaling environment production, and every enterprise will need to do the same. Training agents is becoming one of the largest categories of work in the economy, and the demand cuts across every profession. That is what we are building toward, and it is how we organize human intelligence to power the AI economy. Welcome to Mercor, Deeptune.

OpenAI and Google sell AI models to blacklisted China groups

Get 2 months free with an annual subscription at was $59.88 now $49 . Access to eight surprising articles a day, hand-picked by FT editors. For seamless reading, access content via the FT Edit page on and receive the FT Edit newsletter. Then $75 per month. Complete digital access to quality FT journalism on any device. Cancel or change your plan anytime during your trial. Essential digital access to quality FT journalism on any device. Pay a year upfront and save 20%. Complete digital access to quality FT journalism with expert analysis from industry leaders. Pay a year upfront and save 20%. Check whether you already have access via your university or organisation. Terms & Conditions apply Discover all the plans currently available in your country Digital access for organisations. Includes exclusive features and con

tent. See why over a million readers pay to read the Financial Times.

Show HN: R3 – A Local Code Review Tool for You and Your AI Agent

I built r3 to scratch my own itch when reviewing long design doc and planning doc from AI. The chat interface is inherently bad at tracking multiple pieces of feedback across different parts of a document, so I created a lightweight local web UI to organize the review process.I have been using it on my own projects for one week, and it has made giving structured feedback to AI much easier. I thought others working with AI coding agents might find it useful too.I'd love to hear what you think and...

SHIFT: Survival Prediction from Incomplete and Heterogeneous Genomic Data

Focus to learn more arXiv-issued DOI via DataCite Submission history From: Muhammet Sami Yavuz [ view email ] [v1] Sat, 4 Jul 2026 21:04:57 UTC (2,164 KB) Full-text links: Access Paper: View a PDF of the paper titled SHIFT: Survival Prediction from Incomplete and Heterogeneous Genomic Data, by Muhammet Sami Yavuz and 4 other authors View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs q-bio 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 Connect

ed 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 .

Uncertainty-gated selection for block

Focus to learn more arXiv-issued DOI via DataCite Submission history From: Thomas Rossi [ view email ] [v1] Sat, 4 Jul 2026 20:41:42 UTC (28 KB) Full-text links: Access Paper: View a PDF of the paper titled Uncertainty-gated selection for block-sparse attention, by Thomas Rossi View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs 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 S

mart 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 .

Omni-Sleep: A Sleep Foundation Model via Hierarchical Contrastive Learning of CNS-

Focus to learn more arXiv-issued DOI via DataCite Submission history From: Zhoujie Hou [ view email ] [v1] Sat, 4 Jul 2026 13:45:39 UTC (1,399 KB) Full-text links: Access Paper: View a PDF of the paper titled Omni-Sleep: A Sleep Foundation Model via Hierarchical Contrastive Learning of CNS--ANS Dynamic, by Zhoujie Hou and 5 other authors View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs 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 .

ReCoLoRA: Spectrum-Aware Recursive Consolidation for Continual LLM Fine

Focus to learn more arXiv-issued DOI via DataCite Submission history From: Wentao Lu [ view email ] [v1] Sat, 4 Jul 2026 12:58:26 UTC (72 KB) Full-text links: Access Paper: View a PDF of the paper titled ReCoLoRA: Spectrum-Aware Recursive Consolidation for Continual LLM Fine-Tuning, by Wentao Lu View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs 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? ) Litmap

s 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 .

LLT: Local Linear Transformer for PDE Operator Learning

Focus to learn more arXiv-issued DOI via DataCite Submission history From: Oded Ovadia [ view email ] [v1] Sat, 4 Jul 2026 11:07:09 UTC (9,587 KB) Full-text links: Access Paper: View a PDF of the paper titled LLT: Local Linear Transformer for PDE Operator Learning, by Oded Ovadia 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 math 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 Toggl

e 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 .

长尾胸部X光分类中的亚组漏诊问题研究

近日，一项发表于arXiv的研究聚焦于长尾分布下胸部X光分类的亚组漏诊现象。论文由Ha-Hieu Pham等学者完成，题为“Who Gets Missed in the Tail? Thresholded Subgroup Underdiagnosis in Long-Tailed Chest X-ray Classification”。研究指出，在医学影像分类任务中，数据常呈现长尾分布，导致模型对罕见疾病或特定亚组的诊断性能显著下降，产生系统性漏诊。通过引入阈值化分析方法，工作量化了不同亚组间的诊断差异，揭示了模型在尾部类别上的公平性缺陷。该成果为提升临床AI模型的鲁棒性与公平性提供了重要参考，有助于推动医学影像辅助诊断的可靠部署。 #医学影像 #AI #长尾分布 #胸部X光 #分类 #漏诊 #公平性 #arXiv #研究

时序图网络可解释性新方法

arXiv上的一篇新论文提出了一种针对时序图网络（Temporal Graph Networks）的可解释性方法，名为“内存回溯与拓扑归因”（Memory Backtracking and Topological Attribution）。该方法通过回溯模型的内存状态并分析图拓扑结构，为时序图模型的预测结果提供可解释的归因。研究旨在解决时序图网络在动态图任务中“黑箱”特性的问题，提升模型在社交网络分析、金融交易检测等场景下的可解释性和可信度。 #时序图网络 #可解释性 #内存回溯 #拓扑归因 #机器学习 #人工智能 #论文

最大似然估计的子高斯集中与熵正态性研究

近日，一篇题为“Sub-Gaussian Concentration and Entropic Normality of the Maximum Likelihood Estimator”的学术论文在预印本平台arXiv上发布。该论文由Leighton P. Barnes与Alex Dytso共同撰写，聚焦于最大似然估计（MLE）的理论性质，证明了MLE在特定条件下具有子高斯集中行为以及熵正态性。这一成果为高维统计推断和机器学习算法的收敛性分析提供了新的数学工具。论文自2026年5月首次提交以来，已更新多个版本，受到相关领域研究者关注。 #最大似然估计 #子高斯集中 #熵正态性 #arXiv #统计学习 #信息论 #机器学习 #高维统计