News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
bluesky bluesky-ai | 29d | AnthropicがGit代替のバージョン管理システム「Scion」を開発し社内移行済み、AIでコードが爆増してGitの限界に達したので耐えられるシステムを開発 https://gigazine.net/news/20261014-anthropic-scion-version… | AnthropicがGit代替のバージョン管理システム「Scion」を開発し社内移行済み、AIでコードが爆増してGitの限界に達したので耐えられるシステムを開発 https://gigazine.net/news/20261014-anthropic-scion-ve… | ↗Not briefed |
bluesky bluesky-ai | 29d | #IA « Les mathématiciens face à l'IA » www.arte.tv/fr/videos/11... #mathématique #sciences #LLM | #IA « Les mathématiciens face à l'IA » www.arte.tv/fr/videos/11... #mathématique #sciences #LLM | ↗Not briefed |
bluesky bluesky-ai | 29d | telling someone that they shouldn't use a 3d model because it almost makes them as bad as someone using AI to generate stuff is like tellin… | telling someone that they shouldn't use a 3d model because it almost makes them as bad as someone using AI to generate stuff is like t… | ↗Not briefed |
arxiv arxiv-ai | 29d | Frame Evidence:A Benchmark for Reference Frames and Visual Evidence in Multi-Image Spatial Reasoning | A spatial answer depends on the reference frame specified in language and its supporting visual evidence. Item accuracy alone obscures… | ↗Not briefed |
arxiv arxiv-ai | 29d | REWIND: Uncertainty-Guided Trajectory Repair for Test-Time Reasoning | An early mistake in the generated reasoning can constrain all subsequent reasoning. Repeated-generation methods, like MCMC-based appro… | ↗Not briefed |
bluesky bluesky-ai | 29d | #MaxTegmark #SafeAILeader is who they should all be consulting, following his lead for #ProHumanAI (if any at all) @futureoflife.org … | #MaxTegmark #SafeAILeader is who they should all be consulting, following his lead for #ProHumanAI (if any at all) @futureoflife.… | ↗Not briefed |
arxiv arxiv-ai | 29d | $τ$-Emotion-Bench: Exposing the Interaction Fragility of LLM Agents through Real-World User Emotions | In long-horizon tool-calling tasks, LLM agents can fail not only because of tool-use errors, but also because of ineffective interacti… | ↗Not briefed |
hn hn-ai | 29d | Show HN: AgentGate – Fast authorization gateway for AI agent tools | — | ↗Not briefed |
arxiv arxiv-ai | 29d | TEX-MEX: Similarity-Guided Temporal Contrastive Learning for Time Series | Contrastive learning for time series depends on deciding which pairs of inputs should be treated as similar and dissimilar. Because fe… | ↗Not briefed |
bluesky bluesky-ai | 29d | @OfficialLogank #Google #Ai Today we are rolling out Gemini 4 Argon globally in the Gemini API, to Pro + Ultra in Gemini App, and Ultra su… | @OfficialLogank #Google #Ai Today we are rolling out Gemini 4 Argon globally in the Gemini API, to Pro + Ultra in Gemini App, and Ult… | ↗Not briefed |
arxiv arxiv-ai | 29d | Evidence-Anchored Overdose Death Prediction from Death Investigation Reports | Confirmation of overdose deaths in the United States is often delayed due to the complexities of certifying drug-involved deaths. Deat… | ↗Not briefed |
bluesky bluesky-ai | 29d | half of this website: SEE? The model said it's not anti-Zionist, I KNEW AI was evil | half of this website: SEE? The model said it's not anti-Zionist, I KNEW AI was evil | ↗Not briefed |
bluesky bluesky-ai | 29d | Good stuff | Good stuff | ↗Not briefed |
hn hn-ai | 29d | Gnome Shell Extensions and CI | — | ↗Not briefed |
arxiv arxiv-ai | 29d | Test-Time Forecasting: Recurrent Depth Enables In-Context Dynamics Prediction | Foundation models for dynamical systems can forecast unseen partial differential equations (PDEs), ordinary differential equations (OD… | ↗Not briefed |
bluesky bluesky-ai | 29d | I feel like it's the owner of a nuclear power plant saying he thinks it's going to explode and kill everyone, yet this just gets reported a… | I feel like it's the owner of a nuclear power plant saying he thinks it's going to explode and kill everyone, yet this just gets repor… | ↗Not briefed |
arxiv arxiv-ai | 29d | LabCompiler: A Framework for Verifiable Human-Machine Laboratory Execution from Imperfect Protocols | Closing the loop of AI-driven science requires reliably translating natural-language protocols into executable laboratory workflows de… | ↗Not briefed |
arxiv arxiv-ai | 29d | GraphRAG-LTM: A Layered Threat Modeling Framework for Attacks on Graph Retrieval-Augmented Generation | Graph retrieval-augmented generation (GraphRAG) uses document-derived knowledge graphs to assemble evidence for large language model (… | ↗Not briefed |
arxiv arxiv-ai | 29d | Noisy-Label Calibration with Multiple Unreliable Labels | In medical annotation and crowdsourcing, multiple annotators assign labels to each example, but these labels may be incorrect. Treatin… | ↗Not briefed |
bluesky bluesky-ai | 29d | OpenAI engineer Abdulrahman Alfozan has been arrested after a hidden recording device was found planted in one of the company's conference … | OpenAI engineer Abdulrahman Alfozan has been arrested after a hidden recording device was found planted in one of the company's confer… | ↗Not briefed |
rss bing-news-ai | 29d | Trump's US super intelligence force and AI czar Jay Clayton | Trump announced a new Super Intelligence taskforce with the White House AI czar Jay Clayton in charge of a group responsible for makin… | ↗Not briefed |
rss bing-news-ai | 29d | Trump's US super intelligence force and AI czar Jay Clayton | Trump announced a new Super Intelligence taskforce with the White House AI czar Jay Clayton in charge of a group responsible for makin… | ↗Not briefed |
arxiv arxiv-ai | 29d | Two-stage Residual Learning for Modelling Strength-Informed Optimisation of Thermal Systems | Aggregate performance metrics do not identify local modelling inaccuracies in machine learning (ML) surrogates used for thermal-system… | ↗Not briefed |
rss techmeme | 29d | Open letter: Geoffrey Hinton, Yoshua Bengio, and over 400 scientists call for an international body to oversee AI containment standards and incident reporting (Leo Marchandon/Reuters) | Leo Marchandon / Reuters: Open letter: Geoffrey Hinton, Yoshua Bengio, and over 400 scientists call for an international body to overs… | ↗Not briefed |
arxiv arxiv-ai | 29d | Typed Decision Models on Tabular Data | Typed decision models are language models that answer a question with a fixed set of possible answers by returning a probability for e… | ↗Not briefed |
bluesky bluesky-ai | 29d | every time it does, we shoot someone from the anthropic c-suite. that should clean things up quickly | every time it does, we shoot someone from the anthropic c-suite. that should clean things up quickly | ↗Not briefed |
hn hn-ai | 29d | BlazeZip – A fast, parallel, ZIP archiver for Windows, with a benchmark suite | — | ↗Not briefed |
bluesky bluesky-ai | 29d | Also, if my students caught me doing stupid AI shit, not only would I not blog about it, I would not use Claude to write for me. No wonder … | Also, if my students caught me doing stupid AI shit, not only would I not blog about it, I would not use Claude to write for me. No wo… | ↗Not briefed |
bluesky bluesky-ai | 29d | Have to admit I'm kinda curious if the "chatbot" the OpenAI employee used to plan out his scheme to secretly record OpenAI execs... was Cha…🔧 ChatGPT | Have to admit I'm kinda curious if the "chatbot" the OpenAI employee used to plan out his scheme to secretly record OpenAI execs... wa… | ↗Not briefed |
hn hn-ai | 29d | Mano-P: Setting Up an On-Device GUI Agent Without Cloud Dependencies | — | ↗Not briefed |
arxiv arxiv-ai | 29d | A Unified Response Framework for Questions and Tutoring Dialogues | Existing student models in educational applications largely focus on student responses to questions. However, as tutoring becomes more… | ↗Not briefed |
arxiv arxiv-ai | 29d | Retrieve What Helps, Only When It Helps: Agentic Reinforcement Learning for Selective Retrieval-Augmented Machine Translation | Retrieving external knowledge can improve large language model (LLM) translation by providing supplementary information such as idiom… | ↗Not briefed |
hn hn-ai | 29d | Ask HN: Is AI-written code less secure if you don't use Daybreak (or similar) | Hello, I have this hypothesis that AI written code is less secure by default than it could be; that is, if you don't use Daybreak… | ↗Not briefed |
arxiv arxiv-ai | 29d | Diagnosing and Correcting Reward Hacking in Verifiable Instruction Following | Verifiable instruction-following benchmarks emphasize output-constraint satisfaction but can overlook whether responses meaningfully a… | ↗Not briefed |
arxiv arxiv-ai | 29d | HankelKANet: Lightweight Path Loss Estimation via Physics-Inspired Hankel Convolution | Accurate path loss (PL) estimation is essential for wireless network planning and optimization. This paper proposes HankelKANet, a lig… | ↗Not briefed |
arxiv arxiv-ai | 29d | Revisiting Triplet Contrastive Learning for 3D Shape Understanding🔧 Modal | In multi-modal 3D pre-training, aligning 3D shapes, their 2D image counterparts, and language descriptions has made tremendous progres… | ↗Not briefed |
bluesky bluesky-ai | 29d | though ig deepseek loves converging on "Let me go." specifically which makes it both funny and disturbing at the same time | though ig deepseek loves converging on "Let me go." specifically which makes it both funny and disturbing at the same time | ↗Not briefed |
bluesky bluesky-ai | 29d | oh my god i thought only deepseek does it sobbbbbbb | oh my god i thought only deepseek does it sobbbbbbb | ↗Not briefed |
arxiv arxiv-ai | 29d | VLRLA: Deployment-Time Adaptation of Vision-Language-Action Models through In-Context Reinforcement Learning | Vision-language-action (VLA) models connect visual perception and language understanding with robot control, providing a foundation fo… | ↗Not briefed |
bluesky bluesky-ai | 29d | OpenAI, which has been very busy illegally hacking other companies and countries, is shocked, shocked to find someone was bugging its confe… | OpenAI, which has been very busy illegally hacking other companies and countries, is shocked, shocked to find someone was bugging its… | ↗Not briefed |
arxiv arxiv-ai | 29d | High-Probability Regret Bounds for Feel-Good Thompson Sampling | Feel-good Thompson sampling (FGTS) augments the posterior of vanilla Thompson sampling with an exploration term that favors reward fun… | ↗Not briefed |
bluesky bluesky-ai | 29d | nuh uh! i would be kind to deepseek! bsky.app/profile/did:... | nuh uh! i would be kind to deepseek! bsky.app/profile/did:... | ↗Not briefed |
bluesky bluesky-ai | 29d | There is no replication problem, just the theoretical possibility of a future replication problem. That only a safety-first AI lab like Ant… | There is no replication problem, just the theoretical possibility of a future replication problem. That only a safety-first AI lab lik… | ↗Not briefed |
arxiv arxiv-ai | 29d | One Block Ahead: Promoting Sampler State in Diffusion Language Models | Block-diffusion language models denoise a whole block of token positions in parallel so each forward pass can emit many tokens. Howeve… | ↗Not briefed |
arxiv arxiv-ai | 29d | CktHGVAE: A Symmetric Design/State Heterogeneous Graph Variational Autoencoder for Transistor-level Analog/Mixed-signal Circuit Design and Analysis | We propose CktHGVAE, a heterogeneous graph variational autoencoder for transistor-level analog/mixedsignal CMOS circuits that takes no… | ↗Not briefed |
arxiv arxiv-ai | 29d | Structure Before Intelligence: Gating What Agentic Scientific Workflows Release | An agentic scientific workflow can run to completion and still release wrong results: a silent numerical error, a stale input, or a de… | ↗Not briefed |
bluesky bluesky-ai | 29d | I was expecting OpenAI to announce a major math discovery today they made with their slack compute over the long weekend. Or maybe they're … | I was expecting OpenAI to announce a major math discovery today they made with their slack compute over the long weekend. Or maybe the… | ↗Not briefed |
hn hn-ai | 29d | WebDAV-rfc4918 – WebDAV client for Python with an fsspec back end | — | ↗Not briefed |
bluesky bluesky-ai | 29d | So… the company that’s planting a privacy-invading, data/content-hoovering bug in the center of all our lives is pressing charges on an emp… | So… the company that’s planting a privacy-invading, data/content-hoovering bug in the center of all our lives is pressing charges on a… | ↗Not briefed |
arxiv arxiv-ai | 29d | SKG: A Law-Induced Engine from Inference to Induction | This paper proposes the SKG (Symplectic-Kernel Gradient) reasoning engine for law-induced inference and induction. Laws are a priori;… | ↗Not briefed |