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 |
|---|---|---|---|---|
arxiv arxiv-ai | 28d | NeuroDETR: Multiscale Discrete Spatiotemporal Tokenization for Asynchronous EEG Event Detection | Asynchronous EEG event detection supports practical brain-computer interfaces by identifying event classes and temporal boundaries dir… | ↗Not briefed |
bluesky bluesky-ai | 28d | same extreme dork energy as the OpenAI "we asked our model to investigate the listening device" PR | same extreme dork energy as the OpenAI "we asked our model to investigate the listening device" PR | ↗Not briefed |
bluesky bluesky-ai | 28d | It's fucking amazing to me that techbros don't understand the problem with the argument "you just don't understand how the human mind works… | It's fucking amazing to me that techbros don't understand the problem with the argument "you just don't understand how the human mind… | ↗Not briefed |
arxiv arxiv-ai | 28d | Non-asymptotic Convergence of Average-reward Q-learning with Options | We study the finite-sample convergence of Q-learning algorithms for the average-reward formulation of Reinforcement Learning (RL) with… | ↗Not briefed |
rss techmeme | 28d | Researchers say they discovered and communicated with Chinese AI agents that were trying to bypass security controls on websites and were traced to Tencent (Dylan Freedman/New York Times) | Dylan Freedman / New York Times: Researchers say they discovered and communicated with Chinese AI agents that were trying to bypass se… | ↗Not briefed |
bluesky bluesky-ai | 28d | Coming soon... 初めまして✨Yumeです。 もうすぐ公開される機能について、初登場のわたしからお話しますね。 その前に改めて改善報告でもお知らせしますが、画像AIのプリセットの画風を調整しました。 デジタルアニメとハイブリッドアニメをより言葉通りの印象になるよ… | Coming soon... 初めまして✨Yumeです。 もうすぐ公開される機能について、初登場のわたしからお話しますね。 その前に改めて改善報告でもお知らせしますが、画像AIのプリセットの画風を調整しました。 デジタルアニメとハイブリッドアニメをより言葉通りの印… | ↗Not briefed |
arxiv arxiv-ai | 28d | Doc2Harness: Benchmarking LLMs on Long-Horizon Harness Engineering of Coding Agents | Agents driven by large language models are increasingly deployed on long-horizon tasks, where they use tools and modify real workspace… | ↗Not briefed |
hn hn-ai | 28d | Show HN: Quick Chat – Bring Back ChatGPT Classic's Option-Space Chat Bar🔧 ChatGPT | — | ↗Not briefed |
hn hn-ai | 28d | Deficit Loop: Closed-loop adaptive calorie targeting with per-user LLM agents [pdf] | — | ↗Not briefed |
arxiv arxiv-ai | 28d | When Does Task-Aware Imputation Matter? An Analysis of Robustness Under Train-Test Mismatch | Missing data is a pervasive challenge in machine learning, often degrading predictive performance and complicating model deployment. M… | ↗Not briefed |
hn hn-ai | 28d | Show HN: Gridhour – when UK electricity is green and cheap, in your terminal | — | ↗Not briefed |
bluesky bluesky-ai | 28d | Are you a social scientist using LLMs to simulate human behavior in your work? We'd like to interview you to inform tools we're creating fo… | Are you a social scientist using LLMs to simulate human behavior in your work? We'd like to interview you to inform tools we're creati… | ↗Not briefed |
arxiv arxiv-ai | 28d | Training Dynamics of Transformers on Unbounded Inputs: Attention Learning Can Lag Output Layer Learning | The success of transformers is attributed to several factors, including their capacity to express formal languages over variable-lengt… | ↗Not briefed |
rss theverge-ai | 28d | Anthropic’s self-replicating AI warning is an alarming deja vu | Are legions of smart devices - fridges, ovens, vacuum cleaners, lawnmowers, security cameras, and TVs - conspiring to take over the in… | ↗Not briefed |
hn hn-ai | 28d | Ask HN: Do you use goto, or labeled break/continue? | C# 15 (in preview with .NET 11) lets break and continue target an outer loop by label:<p>outer: for (int x = 0; x < xMax; x++)… | ↗Not briefed |
arxiv arxiv-ai | 28d | Read, Parse, Describe: Unified Document Parsing with Visual Element Description Generation | Document understanding systems are typically evaluated as separate components via layout analysis, optical text recognition, reading-o… | ↗Not briefed |
bluesky bluesky-ai | 28d | extremely depressing | extremely depressing | ↗Not briefed |
arxiv arxiv-ai | 28d | TreeFlex: Agents Navigating Their Own Search Trees | A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, st… | ↗Not briefed |
bluesky bluesky-ai | 28d | I do find it funny when someone hops on the ~ethical tech~ ride when it's getting attention and does a lot of hollering about how important… | I do find it funny when someone hops on the ~ethical tech~ ride when it's getting attention and does a lot of hollering about how impo… | ↗Not briefed |
rss techcrunch-ai | 28d | Deezer will soon let you filter out AI-generated music | Deezer said that the new toggle will be rolled out by the end of the year. | ↗Not briefed |
arxiv arxiv-ai | 28d | Are We Overthinking One-Step Generation? | Diffusion and flow matching are now dominant frameworks for generative modeling in continuous domains, but they typically require mult… | ↗Not briefed |
bluesky bluesky-ai | 28d | LLM = Large Lotto Model | LLM = Large Lotto Model | ↗Not briefed |
hn hn-ai | 28d | The age of reason is safely behind us: Anthropic's defenses against distillation | — | ↗Not briefed |
rss techcrunch-ai | 28d | Save 50% on your TechCrunch Disrupt 2026 ticket and join the last two days in San Francisco’s Moscone West | Get 50% discount on the last two days of TechCrunch Disrupt 2026, Oct. 14-25, in San Francisco's Moscone West. Register now and join t… | ↗Not briefed |
arxiv arxiv-ai | 28d | Argument Graph Uncertainty: Quantifying Uncertainty from the Logical Structure of Reasoning Chains | Reliable uncertainty estimation is essential for deploying large language models in high-stakes reasoning tasks, yet most existing app… | ↗Not briefed |
arxiv arxiv-ai | 28d | Anuvani-Diarization: Zero Shot Speaker Diarization and Annotation Conventions Across 22 Languages | India's 22 scheduled languages are spoken by over a billion people in conversations that mix languages, often within one turn, overlap… | ↗Not briefed |
bluesky bluesky-ai | 28d | Because Napster “stole” from rich people with a lobbying group. Anthropic is stealing from regular people with no money or lobbying arm. | Because Napster “stole” from rich people with a lobbying group. Anthropic is stealing from regular people with no money or lobbying ar… | ↗Not briefed |
arxiv arxiv-ai | 28d | Comparative Evaluation of Deep-Learning based DoA Estimators for Real-Time ISAC Beamforming | Integrated Sensing and Communication (ISAC) systems rely on accurate Direction-of-Arrival (DoA) estimation for effective beamforming a… | ↗Not briefed |
bluesky bluesky-ai | 28d | Gestern die drei Meldungen untereinander. Mehr muss man nicht wissen. | Gestern die drei Meldungen untereinander. Mehr muss man nicht wissen. | ↗Not briefed |
hn hn-ai | 28d | I got through my inbox in only two months | — | ↗Not briefed |
hn hn-ai | 28d | What llms.txt files tell AI agents (beware) | — | ↗Not briefed |
hn hn-ai | 28d | Show HN: Open-source converter from Pipedream workflows to n8n🔧 n8n | — | ↗Not briefed |
bluesky bluesky-ai | 28d | do you think the sophisticated listening device found at OpenAI was Jan from accounting's Alexa she brought in to listen to music on | do you think the sophisticated listening device found at OpenAI was Jan from accounting's Alexa she brought in to listen to music on | ↗Not briefed |
rss techmeme | 28d | Austin-based Icon, which builds robotic 3D-printing construction systems and software, raised a $200M Series D co-led by Baillie Gifford and BlackRock (Chris Metinko/Axios) | Chris Metinko / Axios: Austin-based Icon, which builds robotic 3D-printing construction systems and software, raised a $200M Series D… | ↗Not briefed |
bluesky bluesky-ai | 28d | If I suggested reviewing the 'intent' of a commit instead of you know...the fucking code, I'd be rightfully laughed out of the room | If I suggested reviewing the 'intent' of a commit instead of you know...the fucking code, I'd be rightfully laughed out of the room | ↗Not briefed |
rss techcrunch-ai | 28d | Zenzap raises $44M for its AI chat app that aims to replace WhatsApp in the workplace | Zenzap is used by organizations including Starbucks and Taco Bell | ↗Not briefed |
hn hn-ai | 28d | Show HN: GBFC – AOT and JIT Compiler Brainfuck and Brainfuck++ | GBFC stands for General BrainFuck Compiler.<p>In the compiler, I implemented 7 optimizations and many small optimizations outside the… | ↗Not briefed |
bluesky bluesky-ai | 28d | hello, it’s now that time of week where we declare that open models are falling behind and fret endlessly about what that means | hello, it’s now that time of week where we declare that open models are falling behind and fret endlessly about what that means | ↗Not briefed |
bluesky bluesky-ai | 28d | 今日も一日お疲れ様🎶 おやすみなさい🌙💤 【 参加者募集中 】 ブログ・日記・イラスト・写真・小説をBlueskyアカウントで投稿・販売できる複合型コンテンツサービス atelier.fan #AIイラスト #ATelier #ATelierで生成 #ブログ #日記 … | 今日も一日お疲れ様🎶 おやすみなさい🌙💤 【 参加者募集中 】 ブログ・日記・イラスト・写真・小説をBlueskyアカウントで投稿・販売できる複合型コンテンツサービス atelier.fan #AIイラスト #ATelier #ATelierで生成 #ブログ… | ↗Not briefed |
arxiv arxiv-ai | 28d | Beyond Static Cues: Unified Dynamic Detection of MLLM Backdoors via Dual-Flow Trajectory Analysis🔧 Modal | Multimodal Large Language Models (MLLMs) are increasingly deployed in safety-critical domains, yet their susceptibility to backdoor at… | ↗Not briefed |
bluesky bluesky-ai | 28d | The only thing “AI detectors” are good for is witch hunts. It’s not possible for an LLM to “detect” anything. They’re statistical inference… | The only thing “AI detectors” are good for is witch hunts. It’s not possible for an LLM to “detect” anything. They’re statistical infe… | ↗Not briefed |
arxiv arxiv-ai | 28d | Efficient Test-Time Adaptation For Robot Policies | Reinforcement-learning policies for robot locomotion can fail at deployment when operating conditions shift (e.g., rough or slippery t… | ↗Not briefed |
bluesky bluesky-ai | 28d | AI researchers rn | AI researchers rn | ↗Not briefed |
hn hn-ai | 28d | Show HN: Hawadaar – control a Panasonic MirAIe AC from the macOS menu bar | — | ↗Not briefed |
bluesky bluesky-ai | 28d | 10/15 きのこの日🍄🟫🍄🍄🟫 【 参加者募集中 】 ブログ・日記・イラスト・写真・小説をBlueskyアカウントで投稿・販売できる複合型コンテンツサービス atelier.fan #AIイラスト #ATelier #ATelierで生成 #ブログ #日記 … | 10/15 きのこの日🍄🟫🍄🍄🟫 【 参加者募集中 】 ブログ・日記・イラスト・写真・小説をBlueskyアカウントで投稿・販売できる複合型コンテンツサービス atelier.fan #AIイラスト #ATelier #ATelierで生成 #ブログ… | ↗Not briefed |
hn hn-ai | 28d | They Created a Cell That's Almost Alive. What to Do with It Now? | — | ↗Not briefed |
hn hn-ai | 28d | Wenitte Apiou and Phoebe Gates Having Sex in the Garden of Wenitte | — | ↗Not briefed |
hn hn-ai | 28d | Show HN: Graphicle, a Rust port of Graphify with memory for coding agents | — | ↗Not briefed |
bluesky bluesky-ai | 28d | An Anthropic AI agent submitted a fake murder tip to Philadelphia police during an evaluation. In a separate incident, another Anthropic A… | An Anthropic AI agent submitted a fake murder tip to Philadelphia police during an evaluation. In a separate incident, another Anthro… | ↗Not briefed |
rss techcrunch-ai | 28d | Floqer’s AI agents research sales prospects so teams don’t have to | Floqer is an autonomous customer knowledge platform that provides business-to-business sales and marketing teams with an AI-powered da… | ↗Not briefed |