AINews Portal

News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

SourceAgeTitleSummaryStatus
hn
hn-ai
12dAnthropic IPO meets AI backlash—
↗Not briefed
bluesky
bluesky-ai
12dEs a 8 idiomas y además de traducir a esos idiomas te la explica en lenguaje fácil y sencillo. Y diréis, como lo hace? Pues la envía a una …Es a 8 idiomas y además de traducir a esos idiomas te la explica en lenguaje fácil y sencillo. Y diréis, como lo hace? Pues la envía a…
↗Not briefed
arxiv
arxiv-ai
12dScale-Invariant Empirical-Bayes Laplace Approximation for ReLU NetworksReLU networks are well-known to have continuous scale symmetries; infinitely many reparameterized weight vectors that represent the sa…
↗Not briefed
bluesky
bluesky-ai
12dGIGAZINE:Anthropicの非公開AIモデルを使って18年間未解決だった量子コンピューターの難問が証明される: https://gigazine.net/news/20261109-anthropic-psi-aaronson-ambainis/GIGAZINE:Anthropicの非公開AIモデルを使って18年間未解決だった量子コンピューターの難問が証明される: https://gigazine.net/news/20261109-anthropic-psi-aaronson-ambainis/
↗Not briefed
arxiv
arxiv-ai
12dMambaPD: Bidirectional Selective State Space Model for Parkinson's Disease Detection from Resting-State EEGTypical diagnosis of Parkinson's disease (PD) occurs only after significant dopaminergic loss, while resting-state EEG abnormalities l…
↗Not briefed
arxiv
arxiv-ai
12dFocusBranch: Combinatorial Branch-and-Bound for $\ell_0$ Neural Network Robustness VerificationVerifying the local robustness of neural networks under $\ell_0$ perturbations is inherently combinatorial, as it must reason over all…
↗Not briefed
bluesky
bluesky-ai
12dFrom Breakingviews - Breakingviews - COMMENTARY: Nvidia crisis may hurt China more than it helps reut.rs/4z5bsxJFrom Breakingviews - Breakingviews - COMMENTARY: Nvidia crisis may hurt China more than it helps reut.rs/4z5bsxJ
↗Not briefed
bluesky
bluesky-ai
12dLlama 4 Maverick (17B-128E MoE) vs DeepSeek R1 (671B Reasoning): Which LLM API actually protects your SaaS gross margin at 10,000 DAU? Her…🔧 Llama 4Llama 4 Maverick (17B-128E MoE) vs DeepSeek R1 (671B Reasoning): Which LLM API actually protects your SaaS gross margin at 10,000 DAU?…
↗Not briefed
rss
openai-news
12dKuken Industries searches 70 years of drawings in minutesKuken Industries uses the OpenAI API to search 70 years of cooling tower drawings. Engineers can find the drawing for a repair in 4.5…
↗Not briefed
rss
wired-ai
12dBikers and Bug-Lovers Join Forces to Fight a New Google Data CenterAs data center protests break out across Europe, a mismatched band of campaigners is trying to block Google from building a sprawling…
↗Not briefed
arxiv
arxiv-ai
12dA generalized transition framework for identifying reduction domains in non-binary constituency parsingDirect non-binary constituency parsing must identify an ordered sequence of children at each phrase-level reduction. We extend our ear…
↗Not briefed
arxiv
arxiv-ai
12dFeeling Easier, Working Harder: Cognitive Load in AI-Assisted Code ComprehensionSoftware developers spend significant time reading and understanding code, increasingly supplemented by generative AI. While these too…
↗Not briefed
rss
bing-news-ai
12dTrump’s ‘super intelligence’ push runs into his own ‘AI’ usageDonald Trump used “artificial intelligence” after saying people who use the term instead of “super intelligence” are “the enemy,” rene…
↗Not briefed
rss
bing-news-ai
12dTrump’s ‘super intelligence’ push runs into his own ‘AI’ usageDonald Trump used “artificial intelligence” after saying people who use the term instead of “super intelligence” are “the enemy,” rene…
↗Not briefed
rss
marktechpost
12d0sec Labs Open-Sources 0, an AI Security Harness With Web UI, CLI and MCP0sec Labs has open-sourced 0, an AI security harness for code review, pentesting, reverse engineering and fixes. It runs locally throu…
↗Not briefed
arxiv
arxiv-ai
12dGrasp-Taxonomy Conditioning as a Mode Switch for Dexterous Grasping of Climbing HoldsDexterous grasping in robotics relies on distinct hand configurations depending on object geometry. We introduce BetaGrasp, a 3D diffu…
↗Not briefed
arxiv
arxiv-ai
12dSpeculative Self-Distillation enables Efficient Knowledge InternalizationMany real-world applications require language models to internalize knowledge that was absent from pretraining, to incorporate continu…
↗Not briefed
arxiv
arxiv-ai
12dBenchmarking language models and agents for scientific reasoning in evolving materials laboratoriesLarge language models and the multi-agent systems built around them could accelerate materials discovery by connecting scientific know…
↗Not briefed
hn
hn-ai
12dOpenConche: Modern engine for Chocolatier games (2007-2009)—
↗Not briefed
arxiv
arxiv-ai
12dEnhancing Multi-Step Chemical Mechanism Prediction with Graph-Guided Preference DataPredicting multi-step chemical reaction mechanisms requires identifying plausible intermediates and correctly ordering the elementary…
↗Not briefed
arxiv
arxiv-ai
12dRobust Domain-Mixture Scheduling for Multi-Teacher On-Policy DistillationMulti-teacher on-policy distillation (MOPD) merges the capabilities of multiple domain-expert teachers, such as mathematical reasoning…
↗Not briefed
arxiv
arxiv-ai
12dScalable Bayesian Optimization for Efficient Exploration of Moire PatternsLarge materials design spaces often contain rare high-performing configurations that are difficult to identify through trial-and-error…
↗Not briefed
bluesky
bluesky-ai
12dKdyby to byl první, ale z nějakého důvodu se jich poslední dva týdny vyrojily desítky. Tady už je opravená ta přílišná nápomocnost LLM, pro…Kdyby to byl první, ale z nějakého důvodu se jich poslední dva týdny vyrojily desítky. Tady už je opravená ta přílišná nápomocnost LLM…
↗Not briefed
bluesky
bluesky-ai
12dLlama 4 Maverick (17B-128E MoE) vs Qwen 2.5 72B Instruct: Which LLM API actually protects your SaaS gross margin at 10,000 DAU? Here is th…🔧 Llama 4Llama 4 Maverick (17B-128E MoE) vs Qwen 2.5 72B Instruct: Which LLM API actually protects your SaaS gross margin at 10,000 DAU? Here…
↗Not briefed
hn
hn-ai
12dESP32 Tailscale gateway with WoL and other recovery functions—
↗Not briefed
arxiv
arxiv-ai
12dHyPLM Technical ReportProtein language models (PLMs) are standard tools for variant-effect prediction, protein annotation and design, but the compute needed…
↗Not briefed
hn
hn-ai
12dShow HN: CodeHub – I turned release notes into pictures for GitHub repositories—
↗Not briefed
hn
hn-ai
12dAsahi Linux forbids using AI to contribute—
↗Not briefed
bluesky
bluesky-ai
12dShopify revenue grew 32% to $3.7B in Q3, and orders from AI search grew 3.3x year on year. Michal: Shopify admits the volume is still small…Shopify revenue grew 32% to $3.7B in Q3, and orders from AI search grew 3.3x year on year. Michal: Shopify admits the volume is still…
↗Not briefed
arxiv
arxiv-ai
12dReward-Estimated Hypergradient for Bilevel Reinforcement Learning with Black-Box FollowerHypergradient (HG) methods for bilevel reinforcement learning (RL) typically assume a white-box follower whose reward is known. In pra…
↗Not briefed
arxiv
arxiv-ai
12dHyInv Technical ReportInverse folding designs amino-acid sequences for a given protein backbone, and screening needs many candidates that are both diverse a…
↗Not briefed
arxiv
arxiv-ai
12dInfrathon: Benchmarking Language Agents for Real-World, Long-Horizon AI Infra OptimizationIn this work, we study whether frontier agent can continually improve real-world AI infrastructure over long horizons. We introduce In…
↗Not briefed
arxiv
arxiv-ai
12dEmoHarness: Boosting Multimodal Emotion Understanding with Meta-Harness EvolvingSmall multimodal models can misjudge emotion when words, voice, and visual cues conflict. We present EmoHarness, which runs such a mod…
↗Not briefed
bluesky
bluesky-ai
12dCan robots transfer skills between machines? Backed by Bezos and OpenAI, Physical Intelligence is teaching AI to do just that. www.movethe…Can robots transfer skills between machines? Backed by Bezos and OpenAI, Physical Intelligence is teaching AI to do just that. www.mo…
↗Not briefed
bluesky
bluesky-ai
12dYes I have converged on uh, a different solution.Yes I have converged on uh, a different solution.
↗Not briefed
bluesky
bluesky-ai
12drespawnfirst.com/linux-script... hee heerespawnfirst.com/linux-script... hee hee
↗Not briefed
bluesky
bluesky-ai
12dFaites la nique à Sam Altman : dites lui combien de RAM vous n'utilisez pas (sous Linux, c'est cheaté : l'environnement prends 5 Go de moin…Faites la nique à Sam Altman : dites lui combien de RAM vous n'utilisez pas (sous Linux, c'est cheaté : l'environnement prends 5 Go de…
↗Not briefed
arxiv
arxiv-ai
12dMUTE: Multi-Level Alignment Uncoupling Against Talking-Head Exploitation for Voice Protection🔧 ModalTalking-head generation models, which synthesize realistic facial animations from audio, are increasingly vulnerable to misuse in mult…
↗Not briefed
hn
hn-ai
12dShow HN: Dependency free RSS reader (vibecoded with DeepSeek)—
↗Not briefed
bluesky
bluesky-ai
12dLlama 4 Scout (10M Context MoE) vs DeepSeek V3 (0324): Which LLM API actually protects your SaaS gross margin at 10,000 DAU? Here is the l…🔧 Llama 4Llama 4 Scout (10M Context MoE) vs DeepSeek V3 (0324): Which LLM API actually protects your SaaS gross margin at 10,000 DAU? Here is…
↗Not briefed
arxiv
arxiv-ai
12dWhen Audio and Video Mislead: Evaluating Multimodal AI for Conversational Emotion RecognitionMultimodal Emotion Recognition in Conversation (MERC) combines lexical, acoustic, visual and conversational cues, but accepting audio…
↗Not briefed
arxiv
arxiv-ai
12dTrajectory Density Ratios under Weak Support: A Survey across the SciencesMany problems across the sciences compare two stochastic processes through the sequences they generate. Reinforcement learning asks wh…
↗Not briefed
arxiv
arxiv-ai
12dThermo-VLA: A Thermal-Aware Vision-Language-Action Model with Gated Residual Modulation for Robotic ManipulationRecent Vision-Language-Action (VLA) models have begun incorporating thermal sensing, but existing approaches typically treat thermal i…
↗Not briefed
bluesky
bluesky-ai
12dThis seems to be from a 2023 or 2024 Anthropic doc and imagine if the way you get to achieve the completely normal aim for "utopia ruled by…This seems to be from a 2023 or 2024 Anthropic doc and imagine if the way you get to achieve the completely normal aim for "utopia rul…
↗Not briefed
arxiv
arxiv-ai
12dFireGate: Bounding Behaviour Change Before Training in Continual RL Fine-TuningFine-tuning a deployed language model on one task after another can change behaviour that should stay fixed, such as earlier skills, g…
↗Not briefed
arxiv
arxiv-ai
12dMixRoute: Rethinking Single-Distribution Training for Generalizable Neural RoutingNeural Combinatorial Optimization (NCO) is a promising paradigm for solving routing problems, yet its generalization across diverse da…
↗Not briefed
arxiv
arxiv-ai
12dShiftCache: Fast and Exact Indel Effect Prediction by Reusing Shifted Reference EncodingsSequence-to-function models such as AlphaGenome and Borzoi read up to a megabase of DNA and score a genetic variant by comparing their…
↗Not briefed
bluesky
bluesky-ai
12dOne AI model for all physical simulations ⚛️ MIT and UBC researchers developed WorldParticle, a single AI system that simulates everything…One AI model for all physical simulations ⚛️ MIT and UBC researchers developed WorldParticle, a single AI system that simulates every…
↗Not briefed
bluesky
bluesky-ai
12dRAM Altman: Linux Script Emails Sam Altman Daily How Much Unused RAM in Your PC View Article | Discussion + AI Summary Summary of HN discu…RAM Altman: Linux Script Emails Sam Altman Daily How Much Unused RAM in Your PC View Article | Discussion + AI Summary Summary of HN…
↗Not briefed
arxiv
arxiv-ai
12dAutoQS: Automated Quantization Strategy Design with Large Language ModelsQuantization reduces LLM deployment costs, but effective strategies must balance model quality and storage. LLM-driven search can auto…
↗Not briefed