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 | 1mo | www.theguardian.com/australia-ne.... Soon AI will be communicating with other AI in Government deciding what's best for us all. May as well… | www.theguardian.com/australia-ne.... Soon AI will be communicating with other AI in Government deciding what's best for us all. May as… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Neuro-Memory Fuzzy Inference System for Mimicking Human-like Car Following Behavior | This study presents the Neuro-Memory Fuzzy Inference System (NeMeFIS), a hierarchical machine learning architecture that asymmetricall… | ↗Not briefed |
arxiv arxiv-ai | 1mo | V-CoLA: Vision Token Compression with Linear Attention | Vision-language models (VLMs) have demonstrated impressive capabilities but suffer from substantial computational overhead, as vision… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Les Américains ne vont pas se moquer longtemps de Mistral IA, car l'entreprise française d'intelligence artificielle générative revient en … | Les Américains ne vont pas se moquer longtemps de Mistral IA, car l'entreprise française d'intelligence artificielle générative revien… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Why On-Policy Distillation Sometimes Fails: Vanishing Learning Signals | On-policy distillation (OPD) enables effective capability transfer between language models, yet the mechanisms underlying its failures… | ↗Not briefed |
rss techmeme | 1mo | German search engine Ecosia, used by government services, ditches Mistral for open-source models, including Chinese ones, citing Mistral's disappointing quality (Politico) | Politico: German search engine Ecosia, used by government services, ditches Mistral for open-source models, including Chinese ones, ci… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Read What Matters: Query-Adaptive Quantization for KV Caches | KV-cache entries are stored before their future queries are known, but each decoding query needs precision in different places. We stu… | ↗Not briefed |
rss techmeme | 1mo | Filing: Shanghai-based AI chipmaker Biren raises ~$515M in a new share sale; Biren's Hong Kong-listed stock is down 50%+ from a June peak but up ~70% YTD (Bloomberg) | Bloomberg: Filing: Shanghai-based AI chipmaker Biren raises ~$515M in a new share sale; Biren's Hong Kong-listed stock is down 50%+ fr… | ↗Not briefed |
hn hn-ai | 1mo | Studying metagaming latents in language models | — | ↗Not briefed |
rss techmeme | 1mo | Elon Musk claims "certain oligarchs" are stalling Starlink's India launch to protect a "monopolistic chokehold"; SpaceX signed MOUs with Jio and Airtel in 2025 (Priyanka Salve/CNBC) | Priyanka Salve / CNBC: Elon Musk claims “certain oligarchs” are stalling Starlink's India launch to protect a “monopolistic chokehold”… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Low-Cost Sensor Calibration for Indoor Air Quality Monitoring: A Dataset, Evaluation Scenarios, and a Lightweight Model | Low-cost sensors enable scalable indoor air quality monitoring but require calibration because of nonlinear distortions, noise, and te… | ↗Not briefed |
rss techmeme | 1mo | Sources: Isomorphic Labs, an AI drug discovery startup spun out of Google DeepMind, is in early talks to raise funds at a valuation of at least $40B (Bloomberg) | Bloomberg: Sources: Isomorphic Labs, an AI drug discovery startup spun out of Google DeepMind, is in early talks to raise funds at a v… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Right, so your code harness should enforce a variety of tests like “the LLM must rigorously search for existing open source code before wri… | Right, so your code harness should enforce a variety of tests like “the LLM must rigorously search for existing open source code befor… | ↗Not briefed |
arxiv arxiv-ai | 1mo | MiniVer-V: Identifying Minimal Sufficient Evidence for Short Video Verification🔧 GPT-5 | A core challenge in short-video fact-checking is identifying which evidence is sufficient to support a verification conclusion. Existi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Harness Compilation: Which Decisions Should a Small Vision-Language Model Keep? | Small vision-language models may be able to read external evidence yet struggle to obtain it. We introduce Harness Compilation (HC), a… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SteerCast: Retrieval-Based Latent Steering for Decoder-Only Time Series Forecasting | Time series forecasting aims to predict future values from historical observations and auxiliary features. We propose \textbf{SteerCas… | ↗Not briefed |
bluesky bluesky-ai | 1mo | LLM wikis are interesting forms of arranging information systems, but if semantic search solved sense-making problems we wouldn’t still be … | LLM wikis are interesting forms of arranging information systems, but if semantic search solved sense-making problems we wouldn’t stil… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I think the problem here, from my human information behavior information scientist lens, is that an LLM wiki doesn’t solve the problem of h… | I think the problem here, from my human information behavior information scientist lens, is that an LLM wiki doesn’t solve the problem… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Multimodal Graph Retrieval-Augmented Sequential Recommendation via Collaborative Filtering Paths | Multimodal Large Language Models (MLLMs) have demonstrated strong potential for sequential recommendation through their ability to rea… | ↗Not briefed |
bluesky bluesky-ai | 1mo | AI生成画像やAI生成動画を見分けるGoogleサービス「SynthID Detector」が誰でも使用可能に、Google・OpenAI・AppleのどのAIかも判別可能 https://gigazine.net/news/20261008-synth-id-ai-conte… | AI生成画像やAI生成動画を見分けるGoogleサービス「SynthID Detector」が誰でも使用可能に、Google・OpenAI・AppleのどのAIかも判別可能 https://gigazine.net/news/20261008-synth-id-ai-… | ↗Not briefed |
arxiv arxiv-ai | 1mo | When Lower Reconstruction Loss Hurts: Distributionally Robust Refinement for Low-Bit LLM Quantization | Weight-only post-training quantization (PTQ) relies heavily on reconstruction loss minimization to preserve model quality at low preci… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SafeInferCom: Safe Inference-Time Compute via Verifier-Guided Mid-Generation Intervention for Robotic Task Planning | Large Reasoning Language Models (LRLMs) enable multi-step reasoning for robotic task planning, but continued reasoning can overwrite v… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Cross-species representation learning aligns mouse and human neural dynamics and tracks clinical drug efficacy | Preclinical models poorly predict human drug efficacy, particularly in neurological disorders. Neural activity offers a uniquely rich… | ↗Not briefed |
arxiv arxiv-ai | 1mo | BRACE: Differential Privacy for Dense Associative Memory with LSR Energy | Dense associative memory (DAM) provides an energy-based framework for memory retrieval with close connections to attention mechanisms… | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Lattice of Transition Laws | Diffusion and autoregression (AR) have long been seen as different categories of generative models, with diffusion specialising in con… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Bridging KV-Cache Quantization and Linear Attention: From Theory to Pretrained Weight Migration | KV-cache quantization and linear attention are two representative approaches to tackling the storage and computational costs of Transf… | ↗Not briefed |
arxiv arxiv-ai | 1mo | AliO: Output Alignment Matters in Long-Term Time Series Forecasing | Long-term Time Series Forecasting (LTSF) tasks, which leverage the current data sequence as input to predict the future sequence, have… | ↗Not briefed |
bluesky bluesky-ai | 1mo | GeminiにAIの使われ方を根掘り葉掘り聞いたら、ローカルLLM御用達の業界の一つが選挙戦略を担うコンサルとか事務所Σ(・θ・;) ネガキャンで人々に刺さるキャッチフレーズを考えさせたり、敵対候補を効果的に批判する方法を出させたりするそう。 クラウドだと情報が保持されてい… | GeminiにAIの使われ方を根掘り葉掘り聞いたら、ローカルLLM御用達の業界の一つが選挙戦略を担うコンサルとか事務所Σ(・θ・;) ネガキャンで人々に刺さるキャッチフレーズを考えさせたり、敵対候補を効果的に批判する方法を出させたりするそう。 クラウドだと情報が保… | ↗Not briefed |
hn hn-ai | 1mo | An Upper Bound of 9/4 for the Matrix Multiplication Exponent [pdf] | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Do Flatter Minima Drive Better Generalization? An Algorithmic Separation in Grokking | Flat loss landscapes have long been linked to better generalization in neural networks. However, its role as a causal mechanism for ge… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Dissecting Representation Structure in Vision Transformers: A Rigorous Architectural Study | Representation structure is crucial for understanding Vision Transformer (ViT) architectures and their generalization behavior. Howeve… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PageWeaver: KV-Guided Query Unions for Sparse Attention | Dynamic sparse attention limits the KV pages selected by each query, but a small support does not necessarily yield efficient GPU work… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This guy's whole channel is "Can I AI Slop My Way Through Projects That Could Be Interesting But Aren't Because I Don't Care and Just Tell … | This guy's whole channel is "Can I AI Slop My Way Through Projects That Could Be Interesting But Aren't Because I Don't Care and Just… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Selective Listening: Mechanism-Guided Control of Audio Influence in Large Audio-Language Models | Large audio-language models (LALMs) exploit multimodal evidence, yet task-irrelevant audio can alter text-reasoning decisions when lis… | ↗Not briefed |
bluesky bluesky-ai | 1mo | llm life hack: "what if we consider this a state machine" | llm life hack: "what if we consider this a state machine" | ↗Not briefed |
bluesky bluesky-ai | 1mo | specifically they made a really obvious and boneheaded scheduling error in talking to opposing counsel, and when that was pointed out the r… | specifically they made a really obvious and boneheaded scheduling error in talking to opposing counsel, and when that was pointed out… | ↗Not briefed |
hn hn-ai | 1mo | Self-Modeling Interventions Modulate Emergent Misalignment | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | courrier.jp/news/archive... この馬鹿の態度どっかで見たなと思ったら王権神授説だ。「神様が王権授けたんだから王様は何やってもいいんだ」という理屈。 | courrier.jp/news/archive... この馬鹿の態度どっかで見たなと思ったら王権神授説だ。「神様が王権授けたんだから王様は何やってもいいんだ」という理屈。 | ↗Not briefed |
bluesky bluesky-ai | 1mo | I’ve worked at an org that did that in the pre-LLM days. Most of the work I did for them was helping them effectively undo that decision (t… | I’ve worked at an org that did that in the pre-LLM days. Most of the work I did for them was helping them effectively undo that decisi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | What to Admit and How to Present: Governing Persistent Memory in LLM Agents | Persistent memory can improve personalization in LLM agents but can also induce sycophancy and cross-domain leakage. We distinguish tw… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Predictive Multiplicity in Cell-Fate Assignment: Label-Free Rashomon Sets and the Limits of Per-Cell Certification | Single-cell trajectory inference maps transcriptomic measurements onto developmental continua, yet configurations that fit the data eq… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Front page in the print newspaper today. > “I’m confident that even should the upper administration in the CU system decide they don’t wan… | Front page in the print newspaper today. > “I’m confident that even should the upper administration in the CU system decide they don’… | ↗Not briefed |
arxiv arxiv-ai | 1mo | RAG-Stress: Probing the Limits of Evidence Reliance in Retrieval-Augmented Generation | Following retrieved evidence does not guarantee factual correctness: misleading evidence can induce a model to replace an answer it pr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | If the Albanese government wants AI developers to demonstrate their safety systems are effective "under proposed laws drawing on banking an… | If the Albanese government wants AI developers to demonstrate their safety systems are effective "under proposed laws drawing on banki… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Higher-Order Action Supervision Makes A Strong Policy Class | Modern data-driven decision-making methods, such as imitation learning (IL) and reinforcement learning (RL), have achieved great succe… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It's a real assertion, from a real person, argued in a real blog post. 🫠 👇 | It's a real assertion, from a real person, argued in a real blog post. 🫠 👇 | ↗Not briefed |
bluesky bluesky-ai | 1mo | All mathematical areas need new frameworks and they have to decide which problems are interesting. OpenAI doesn’t have sufficient experts … | All mathematical areas need new frameworks and they have to decide which problems are interesting. OpenAI doesn’t have sufficient exp… | ↗Not briefed |
arxiv arxiv-ai | 1mo | VAMR: Multi-Question Agentic Reasoning for Efficient Long-Form Video Understanding | Long-form video understanding often involves multiple questions about different aspects of the same recording. Yet existing video agen… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Has anyone considered that “integrating” Claude or another LLM as enterprise software is really just paying for corporate espionage | Has anyone considered that “integrating” Claude or another LLM as enterprise software is really just paying for corporate espionage | ↗Not briefed |
bluesky bluesky-ai | 1mo | Background [no paywall] www.nytimes.com/2026/10/07/o... | Background [no paywall] www.nytimes.com/2026/10/07/o... | ↗Not briefed |