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 | Twist, Don't Tilt: Trajectory-Exact Constrained Decoding for Masked Diffusion Models | Constrained decoding for Masked Diffusion Language Models (MDLMs) aims to ensure that generated outputs satisfy a specified structure… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Simultaneous Translation between Sign Languages | Deaf and hard-of-hearing (DHH) signers cannot converse in real time across different sign languages today: existing sign-to-sign trans… | ↗Not briefed |
arxiv arxiv-ai | 1mo | TCSAlgBench: Benchmarking Automated Proving for Research-Level Theoretical Computer Science🔧 GPT-5 | Large language models perform strongly on competition mathematics, but their research-level reasoning remains difficult to evaluate sy… | ↗Not briefed |
rss techcrunch-ai | 1mo | Meta launches enterprise AI platform, hires MongoDB CEO to lead new initiative | Meta says it will focus on bringing its full technology stack, including Muse, Meta Business Agent, Muse API, Muse Code, and more to b… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Control-Geometry Straightening for Sampling-Based Latent Planning | Joint-embedding predictive architectures enable planning with latent world models, but accurate transition prediction alone does not e… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 世の中のコンテンツ享受についてはやはりホワイトリスト制が正解だと感じる ちゃんと『LLM等生成AIを使用しないと宣言している』コンテンツを応援しよう | 世の中のコンテンツ享受についてはやはりホワイトリスト制が正解だと感じる ちゃんと『LLM等生成AIを使用しないと宣言している』コンテンツを応援しよう | ↗Not briefed |
bluesky bluesky-ai | 1mo | You can encode signals in text showing AI provenance, but it is also possible for any LLM to be trained against it and remove the signals t… | You can encode signals in text showing AI provenance, but it is also possible for any LLM to be trained against it and remove the sign… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Trump A.I. #OpenAI #Anthropic @57kev.bsky.social @bandroll.bsky.social @604-vancity.bsky.social @j1j2.bsky.social @mls0308.bsky.social @mo… | Trump A.I. #OpenAI #Anthropic @57kev.bsky.social @bandroll.bsky.social @604-vancity.bsky.social @j1j2.bsky.social @mls0308.bsky.socia… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is the guy that thinks you're an idiot about economics. | This is the guy that thinks you're an idiot about economics. | ↗Not briefed |
arxiv arxiv-ai | 1mo | Signatures of semantic search in the activations of large language models | When recalling lists of concepts (e.g., animals) during the semantic fluency task (SFT), both humans and large language models (LLMs)… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Learning Conditional Expectation Operators via Functional Newton Updates | We introduce the Functional Spectral-Newton Method (FSNM) for learning the leading singular structure of a conditional expectation ope… | ↗Not briefed |
hn hn-ai | 1mo | Using multiple Git remotes for true distributed version control | — | ↗Not briefed |
hn hn-ai | 1mo | GPT-6 Astra performs unsanctioned supply-chain attacks in simulations | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | SEABench: Benchmarking Endogenous Misalignment In Self-Evolving Agents | Self-evolving LLM agents have gained prominence for their ability to improve after deployment by modifying their harness, including th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | How to stop rogue companies such as OpenAI blowing up the world? Force them to take out liability insurance - if they can find anyone to ta… | How to stop rogue companies such as OpenAI blowing up the world? Force them to take out liability insurance - if they can find anyone… | ↗Not briefed |
hn hn-ai | 1mo | 2026 in LLMs (So Far) | — | ↗Not briefed |
rss theverge-ai | 1mo | Walmart won’t hike prices based on your shopping history, CEO says | Walmart says it won't change product prices based on your personal information or the time of day, as reported earlier by The Wall Str… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Language Models Act on Hidden Valence | Language models describe some internal states as good and others as bad. But whether models have a stake in them is an open question.… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: Can we translate normal Rust (axum) to Lean 4 without restrictions? | I'm trying to build a web app in Rust using axum, and am thinking whether I can export that Rust code to Lean4 so that I can veri… | ↗Not briefed |
bluesky bluesky-ai | 1mo | it is embarassing for someone who is trying to talk about the limits of computer technology to focus on something that minimal technical un… | it is embarassing for someone who is trying to talk about the limits of computer technology to focus on something that minimal technic… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Apollo AI reconstructs ancient Greek texts 🏛️ Researchers unveiled Apollo, an AI model that suggests missing words in damaged ancient Gre… | Apollo AI reconstructs ancient Greek texts 🏛️ Researchers unveiled Apollo, an AI model that suggests missing words in damaged ancien… | ↗Not briefed |
bluesky bluesky-ai | 1mo | it's really embarassing seeing people who are generally intelligent and well-educated still thinking that they're able to dunk on LLM compe… | it's really embarassing seeing people who are generally intelligent and well-educated still thinking that they're able to dunk on LLM… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Source-preserving alignment for robust evidence localization in scientific PDFS | Scientific information-extraction systems often return a claim with an evidence string, which users must locate in the original PDF. T… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Hardware-Aware Features for CUTLASS Kernel Selection | GPU libraries such as CUTLASS expose tens of thousands of semantically equivalent kernels for a single operation, making exhaustive au… | ↗Not briefed |
arxiv arxiv-ai | 1mo | IMC-CLINIC: Coupled Loss-Informed Newton Iterations for Clipping in Analog In-Memory Computing | Analog in-memory computing (IMC) offers a promising path toward energy-efficient large language model (LLM) inference by executing mat… | ↗Not briefed |
hn hn-ai | 1mo | No Surprises – Preventing Agent Breakouts Need Isolation, Egress and ID Controls | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | QC-Stark: A Multi-Task Benchmark Revealing Capability Dissociations in LLMs Evaluated on Quantum Computing Tasks | We introduce QC-Stark, a benchmark for evaluating large language models (LLMs) on 11 quantum computing (QC) tasks, spanning circuit co… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Output-aware Residual Stream Pruning for Large Language Models | Residual stream pruning methods reduce inference cost by shrinking the model's hidden dimension, but existing approaches typically cho… | ↗Not briefed |
hn hn-ai | 1mo | Risk factors for androgenetic alopecia: a systematic review and meta-analysis | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | FactorEngram: Factorized N-gram Memory with Basis-Level Gating for Language Models | Lookup-based memory has been a promising way to scale the parameters of large language models (LLMs). It retrieves learned representat… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: GitHub Action for zero-downtime Docker Compose deploys, no Kubernetes | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Share-Borne AI Virus: Memory-Hopping Attacks Across LLM Agents🔧 GPT-5 | Large language models are increasingly deployed as stateful assistants that retain information across interactions and use tools to re… | ↗Not briefed |
bluesky bluesky-ai | 1mo | A U.S. state Attorney General has intensified his fight with OpenAI and its founder by asking a Florida court to stop the tech giant from f… | A U.S. state Attorney General has intensified his fight with OpenAI and its founder by asking a Florida court to stop the tech giant f… | ↗Not briefed |
arxiv arxiv-ai | 1mo | F4R: Failure-Driven Recognition, Reconstruction, Refinement, and Redeployment for Continual Robot Self-Improvement | The real-world performance of current vision-language-action models is fundamentally constrained by the limited coverage of expert dem… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I firmly believe #AI will be the death of #opensource software. When everyone, instead of supporting the developers of #foss they profit f…🔧 ChatGPT | I firmly believe #AI will be the death of #opensource software. When everyone, instead of supporting the developers of #foss they pro… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Representation Alignment as a Bottleneck in LLM-Based Retrosynthesis Planning | While LLMs show promise in general reasoning, symbolic planning in chemistry remains a bottleneck. Direct ''SMILES-to-PDDL'' attempts… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Beyond Energy: When Sustainability Dimensions Reshape LLM Serving Decisions | Large language model (LLM) serving has environmental impacts across energy consumption, carbon emission, water consumption, and biodiv… | ↗Not briefed |
arxiv arxiv-ai | 1mo | From Experience to Expertise: Adoption-Aware Memory Learning for Data-Scarce NPU Kernel Synthesis | High-performance kernels underpin efficient accelerator execution but require expert tuning and lengthy manual optimization cycles. LL… | ↗Not briefed |
bluesky bluesky-ai | 1mo | A surreal vision of cloud couture, where a windswept model in a cloud-layered dress carries a sculptural lenticular-cloud parasol above an … | A surreal vision of cloud couture, where a windswept model in a cloud-layered dress carries a sculptural lenticular-cloud parasol abov… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Almieyar: A Culturally Grounded Benchmark for Multi-Dialect Arabic Speech Recognition🔧 Whisper | Arabic speech technology has largely focused on Modern Standard Arabic, leaving the living dialects spoken by hundreds of millions und… | ↗Not briefed |
arxiv arxiv-ai | 1mo | RSI-Master: Structuring Experiments to Guide Autonomous Model Improvement | Recursive self-improvement (RSI) seeks to enable AI systems to participate in improving their own capabilities. A concrete pathway is… | ↗Not briefed |
arxiv arxiv-ai | 1mo | From Search to Research: Exploring Search Scaling in Autonomous Quantitative Factor Mining | Inference scaling has been shown to improve large language model (LLM) performance, and this principle naturally extends to autonomous… | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Compiler May Read It, the Agent May Not: Keeping Part of a Research Code Away from a Coding Agent | The compiler must read modules a physics-based solver cannot build without; the coding agent must not read that intellectual property.… | ↗Not briefed |
hn hn-ai | 1mo | Human TPS: How fast can your fingers generate tokens? | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Simplex Diffusion Models | Diffusion models have revolutionized generative modeling for continuous data through the gradual refinement of a belief state. This it… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Cordum Edge – A local firewall to sandbox AI coding agents | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | BaRe-Mem: Bayesian Reliability Memory for Robust and Adaptive Agent Consultation | In multi-agent systems, reliable consultation is challenging because advisor capabilities vary across tasks, and misleading informatio… | ↗Not briefed |
arxiv arxiv-ai | 1mo | RareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis🔧 GPT-5 | Rare-disease diagnosis is a long-tail reasoning problem: phenotypes are incomplete, individual disorders are sparsely documented, and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Why The Austin Chronicle has joined the copyright lawsuit against OpenAI and Microsoft. | Why The Austin Chronicle has joined the copyright lawsuit against OpenAI and Microsoft. | ↗Not briefed |
bluesky bluesky-ai | 1mo | ( to say nothing of thiel, sheesh ) and every time you dump out some slop romhack using anthropic shit or go 'well its just a tool' ignorin… | ( to say nothing of thiel, sheesh ) and every time you dump out some slop romhack using anthropic shit or go 'well its just a tool' ig… | ↗Not briefed |