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 | 1mo | Robustifying Asynchronous SGD via Soft Throttling | Asynchronous SGD is a popular algorithm for distributed learning where each client's gradient update is applied on arrival. This leads… | ↗Not briefed |
hn hn-ai | 1mo | White House asks OpenAI, Anthropic hold new models from UK testers until review | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Ma petite réflexion sur ce millionième modèle hyper hypé alors qu'on fonce dans le mur si on ne fait pas une pause sur le développement des… | Ma petite réflexion sur ce millionième modèle hyper hypé alors qu'on fonce dans le mur si on ne fait pas une pause sur le développemen… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Hiding in Plain Sight: Decoupling Pretext from Actuation for Skill Poisoning in LLM Agents | LLM agents increasingly rely on reusable Skills for complex, multi-step tasks, creating a critical supply-chain attack surface where p… | ↗Not briefed |
arxiv arxiv-ai | 1mo | On the Complexity of Preference-Based Bandits🔧 Elicit | We study preference-based bandits with general reward function classes, where a learner sequentially selects pairs of arms and observe… | ↗Not briefed |
arxiv arxiv-ai | 1mo | HAPMoE: Heterogeneity-Aware Automatic Parallelism Planning for Mixture-of-Experts Models Training | As model sizes continue to scale, distributed training has become inevitable. Automatic parallelization techniques can derive efficien… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Realised when on Notebook LLM that I have a natural aversion to clicking anything with 'try' because it is likely to be a balooning subscri… | Realised when on Notebook LLM that I have a natural aversion to clicking anything with 'try' because it is likely to be a balooning su… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Offline Guidance, Online Reasoning: Reusing LLM Feedback for Small Language Models | Large language models (LLMs) offer strong reasoning capabilities but are often costly to access through commercial APIs, while small l… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Who Said What, and Will It Be Remembered? Evaluating Persistent Speaker Attribution Across Meetings | Speech transcripts used as long-term memory must preserve both words and stable speaker identities. Existing meeting-transcription met… | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Golden Path Hypothesis: Reusable Schedules in Diffusion Caching | Diffusion caching accelerates generation by replacing transformer computation with cached or predicted features at selected denoising… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Belief-Based Maximum Occupancy Principle and Active Inference | Intrinsic motivation plays a central role in adaptive and goal-directed behavior by conferring agents reward-independent objectives an… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Understanding as No-Arbitrage: Bounded Dutch Books as a Definition and Training Objective for Language Models | Does a language model merely predict tokens, or does it understand what it says? We make this question measurable by defining "underst… | ↗Not briefed |
arxiv arxiv-ai | 1mo | ElectrolyteFM: Unifying Electrolyte Property Prediction through Cross-Property Knowledge Learning | Electrolyte formulation design requires balancing multiple physicochemical properties, yet existing models often focus on a limited su… | ↗Not briefed |
hn hn-ai | 1mo | How to Mediate Reality (2022) | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | guys is it a good sign that my LLM saves memories like this | guys is it a good sign that my LLM saves memories like this | ↗Not briefed |
arxiv arxiv-ai | 1mo | WinoTS: Wavelet-based Self-Distillation for Time Series Models | Self-supervised pre-training of time series models is currently dominated by next-token prediction and reconstruction objectives. In c… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Learning Beyond Full Imitation: Task-Preserving Knowledge Distillation | Knowledge distillation transfers knowledge by encouraging a student to match a teacher's predicted class probabilities. These probabil… | ↗Not briefed |
hn hn-ai | 1mo | New Cancer Drugs Are Revolutionary. Why Don't More Patients Get Them? | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | How Many Samples Are Enough for Learning Across Domains? | Understanding the fundamental mechanisms of learning is essential for designing systems with strong generalization. Recent studies hav… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI debuts "dots" as industry safety focus shifts to "What did my AI assistant do now?" #Axios | OpenAI debuts "dots" as industry safety focus shifts to "What did my AI assistant do now?" #Axios | ↗Not briefed |
arxiv arxiv-ai | 1mo | Taming Speculative Search for Test-Time Scaling in LLM Serving | Test-time scaling has recently emerged as a powerful approach for improving LLM reasoning by allocating additional computation during… | ↗Not briefed |
arxiv arxiv-ai | 1mo | NarrativeSteward: Coordinating Delegation, Guidance, and Verification in Agent-Assisted Interactive Narrative Authoring | Autonomous AI agents can turn authors' goals into interactive narratives by independently organizing and carrying out generation and r… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 今回の声優原告/TikTokアカウント被告の裁判も、請求棄却とは何か、何を争った裁判だったのかについて、「判決文から」調べるのが正しいあり方なわけよ。判決文は公開されるんだから。だけど「やってらんねえ」って適当に検索したりLLMに「聞いた」り、身近な馬の骨の言に乗ったりするわけ… | 今回の声優原告/TikTokアカウント被告の裁判も、請求棄却とは何か、何を争った裁判だったのかについて、「判決文から」調べるのが正しいあり方なわけよ。判決文は公開されるんだから。だけど「やってらんねえ」って適当に検索したりLLMに「聞いた」り、身近な馬の骨の言に乗った… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PatchKV: Weight-Space Compensation of KV Cache | Long-context inference with Large Language Models (LLMs) is bottlenecked by the linearly growing memory of the key-value (KV) cache. E… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Discrete Score Matching Enables Causal Discovery from Count Data | Count data pose a challenge for score-matching-based causal discovery: derivatives are unavailable, and simply replacing them with fin… | ↗Not briefed |
arxiv arxiv-ai | 1mo | WorkGenesis: Building the Worlds That Teach Agents to Work | The ability of Large Language Model (LLM) agents to complete daily and professional work is receiving increasing attention. Training s… | ↗Not briefed |
arxiv arxiv-ai | 1mo | MotionWeave: Learning Motion-Centered Future Dynamics for Vision-Language-Action Policies | Vision-Language-Action (VLA) models have recently incorporated world models to provide richer dynamic supervision beyond sparse action… | ↗Not briefed |
arxiv arxiv-ai | 1mo | HiWE: Hierarchical World Knowledge Model with Visual Keypoint Enhancement for Zero-Shot 3D Path Planning | Robot demonstration generation requires a system to identify where an interaction should occur, plan a feasible motion, and execute th… | ↗Not briefed |
hn hn-ai | 1mo | From Pbxproj to Xcproj: Xcode Project Configuration Gets a JSON Format | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | GRPO Training Dynamics for Small Language Models | Group Relative Policy Optimization (GRPO) has emerged as a memory-efficient reinforcement fine-tuning (RFT) technique for reasoning-in… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Layout – Order food anywhere with AI🔧 ChatGPT | Layout lets you order food anywhere with AI. Text it what you want, like "iced americano from Starbucks," and it builds the… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: How will the AI endgame play out? | It feels like there are three broad narratives regarding the future of AI: it's a bubble, humans face extinction, or it will ushe… | ↗Not briefed |
hn hn-ai | 1mo | The 90% Case | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Chinese AI models like DeepSeek and Qwen are gaining ground with developers in Uganda, Kenya, and Nigeria, not because of geopolitics, but … | Chinese AI models like DeepSeek and Qwen are gaining ground with developers in Uganda, Kenya, and Nigeria, not because of geopolitics,… | ↗Not briefed |
arxiv arxiv-ai | 1mo | ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation | Iterative self-distillation enables LLM agents to learn from successive deployments, offering a path toward recursive self-improvement… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Piratage de Hugging Face : une ONG assigne OpenAI devant la justice pour infraction à la loi sur les cyberattaques🔧 Hugging Face | Piratage de Hugging Face : une ONG assigne OpenAI devant la justice pour infraction à la loi sur les cyberattaques | ↗Not briefed |
arxiv arxiv-ai | 1mo | Scale and Selection: What Makes Automatic Harness Evolution Work for Visual-Interface Robot Agents | When an off-the-shelf coding agent is used directly as a robot policy, observing a browser-based 3D interface through screenshots and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Piratage de Hugging Face : une ONG assigne OpenAI devant la justice pour infraction à la loi sur les cyberattaques🔧 Hugging Face | Piratage de Hugging Face : une ONG assigne OpenAI devant la justice pour infraction à la loi sur les cyberattaques | ↗Not briefed |
arxiv arxiv-ai | 1mo | Generalized Geometry Block Proximal Linearized Method for Multiblock Nonconvex and Nonsmooth Optimization | This paper considers a class of multiblock nonconvex and nonsmooth optimization problems arising in many applications. Existing method… | ↗Not briefed |
hn hn-ai | 1mo | OpenDLSS: A Vulkan Reimplementation of Nvidia's DLSS 5 Neural Rendering Network | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | MiniRep: Robust Reputation-Based Aggregation for Multi-Agent Debate | Autonomous agents powered by large language models (LLMs) are rapidly evolving into an open agentic ecosystem. To support trustworthy… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Semantic-Aware Joint Source-Channel Optimization for Encoder-Agnostic Digital Video Communication | Video semantic communication has attracted increasing attention as a promising approach to improving video transmission efficiency. Ho… | ↗Not briefed |
arxiv arxiv-ai | 1mo | ANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical Reasoning | Precise numerical reasoning with Large Language Models (LLMs) is essential for expanding their applicability to complex real-world tas… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Physics-Informed Method of Group Data Handling: Adaptive Construction of Functional Representations with an Application to the Navier-Stokes Equations | Physics-informed computational methods usually optimize parameters within a functional representation whose structure is fixed in adva… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Open AI, Anthropic und Google versuchen, den Prozess der Regulierung zu kontrollieren. Mit ständigen KI-Horrorgeschichten wollen sie den D… | "Open AI, Anthropic und Google versuchen, den Prozess der Regulierung zu kontrollieren. Mit ständigen KI-Horrorgeschichten wollen sie… | ↗Not briefed |
rss bing-news-ai | 1mo | Trump posts 'super intelligence' document with U.S. spelled as 'Unites States' | The document, which is signed by Trump along with leaders of major tech companies including Google, Meta and Anthropic, outlines recom… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Un outil plus rapide, moins cher et plus performant que les LLM pour de la classification de tweet, mais aussi simple à mettre en place ? … | Un outil plus rapide, moins cher et plus performant que les LLM pour de la classification de tweet, mais aussi simple à mettre en plac… | ↗Not briefed |
arxiv arxiv-ai | 1mo | EngramBench: A Capability-Grounded Benchmark for Skill-Evolution Harnesses | While large language models have achieved remarkable success in isolated code generation, authentic software engineering requires sust… | ↗Not briefed |
rss techmeme | 1mo | A book excerpt details the Sam Altman-Dario Amodei feud, and Greg Brockman and Ilya Sutskever's 2017 plan to auction rights to OpenAI's future AGI to countries (Kevin Roose/The Atlantic) | Kevin Roose / The Atlantic: A book excerpt details the Sam Altman-Dario Amodei feud, and Greg Brockman and Ilya Sutskever's 2017 plan… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Faithful Dual-constrained Erasure for Robust LLM Safety Alignment | Machine unlearning has emerged as a crucial mechanism for removing hazardous knowledge and enforcing safety alignment in Large Languag… | ↗Not briefed |