fx is now available in the AI SDK harness layerIntroducing wrapturePolimill builds Japan's next-generation public AI infrastructurev2.1.252rust-v0.152.0-alpha.7.2Time Capsule of Testable Human Knowledge: 41 Years of Jeopardy! in a Single Free Local ModelSet per-user budgets on AI GatewayA milestone in expanding access to AIRating the Raters: Rasch Measurement Theory for LLM EvaluationHow our agents build on-brand pages with design.mdNot All Explanations Are Sought: Information-Seeking Psychology for Human-Centered XAIRetrieving Relations, Detecting Fallacies: A RAG Approach to Political Debate AnalysisLLM-Augmented Causal Discovery: Probabilistic Fusion of Edge Existence and OrientationHypothesize, Evaluate, Refine: A Scientific Agent for PDE Discovery with Unknown Spatial Coefficient Fieldsfx is now available in the AI SDK harness layer

Introducing wrapture New from Graham Dumpleton (of wrapt , mod_wsgi, and New Relic’s Python agent fame), who describes Wrapture as taking the monkeypatching ideas from wrapt and extending them to apply to testing and tracing at the same time. Wrapture ( full documentation here ) makes it easy to wra

Polimill uses OpenAI GPT models and Codex to help municipalities search and use administrative knowledge while accelerating development.

What’s changed Fixed Bash commands failing with “task output swap refused (tasks dir moved or linked)” on some Macs Fixed “always allow” not saving in a project that has no .claude/settings.local.json yet Fixed Remote Control sessions hosted by Claude Desktop or VS Code stalling for minutes after a

近期更新：rust-v0.152.0-alpha.7.2；rust-v0.152.0-alpha.7.1；0.152.0-alpha.7；0.152.0-alpha.6；0.152.0-alpha.5。Release 0.152.0-alpha.7.2

arXiv:2608.27459v1 Announce Type: new Abstract: In 2011, IBM’s Watson was something like a sealed capsule of its era’s queryable knowledge. Its DeepQA system defeated the strongest human Jeopardy! champions, but the knowledge that let it do so lived in a curated billion-document corpus running on a

You can now use AI Gateway budgets to set a dollar spending limit for each user on your team. The limit covers spend from every API key attributed to the user, along with their app tokens. After it’s reached, AI Gateway rejects new requests until the budget resets or is increased. This is useful for

ChatGPT Ads reaches $1 billion in annualized revenue run rate and expands globally, supporting broader access to AI through free and affordable options.

arXiv:2608.27463v1 Announce Type: new Abstract: LLMs now sit on every side of evaluation: as examinees scored on benchmarks, judges of other models’ outputs, and raters of human-generated content. Each paradigm can be viewed as a measurement problem, where a latent property of an object is probed wi

Across Vercel, we use coding agents to design and build pages that have to look and feel like Vercel. The typography, color, and composition all need to carry the same judgment that we put into the pages we already ship ourselves. We recently wrote about product-design , our skill that teaches agent

arXiv:2608.27464v1 Announce Type: new Abstract: This position paper argues that human-centered explainable AI (HCXAI) should incorporate insights from the psychology of information seeking. Drawing on Sharot and Sunstein’s framework of information-seeking motives, we propose that people evaluate whe

arXiv:2608.27471v1 Announce Type: new Abstract: Fallacies are arguments that employ invalid reasoning, making their automatic detection critical in sensitive contexts such as high-stakes political debates, where public opinion is shaped. Spotting a fallacious argument requires contextual knowledge b

arXiv:2608.27472v1 Announce Type: new Abstract: Bayesian network structure learning (BNSL) from observational data struggles with orientation identifiability, while large language models (LLMs) offer broad but often unreliable causal knowledge. We propose combining these complementary sources throug

arXiv:2608.27475v1 Announce Type: new Abstract: Discovering PDEs in heterogeneous media requires jointly identifying the governing operator and the unknown spatial fields that parameterize it. These tasks are coupled: changing field placement changes the differential law, while a sufficiently flexib