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 | 29d | Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes | City-scale 3D grounding is commonly formulated as candidate ranking, which requires precomputed instance proposals and becomes ineffic… | ↗Not briefed |
arxiv arxiv-ai | 29d | ABot-ER: Unified Embodied Reasoning through Agentic Data-Model Co-Evolution | We introduce \textbf{ABot Embodied Reasoning (ABot-ER)}, a unified embodied reasoning foundation model that breaks the tension between… | ↗Not briefed |
arxiv arxiv-ai | 29d | Choosing LLMs for Strategic Agent Interactions | Game-theoretic benchmarks claim that large language model (LLM) strategic decision-making varies across game types, interaction struct… | ↗Not briefed |
arxiv arxiv-ai | 29d | Let the Mind Conceive, Let the Hand Achieve: Following Multimodal Instructions for Long-Horizon Robotic Manipulation | A robot following a multi-step instruction needs to act on the right objects in the right order, even as the scene changes. It must al… | ↗Not briefed |
bluesky bluesky-ai | 29d | Yes. LLM companies do not care much about the license of the software they have in their training data because their training data is close… | Yes. LLM companies do not care much about the license of the software they have in their training data because their training data is… | ↗Not briefed |
arxiv arxiv-ai | 29d | Lightweight but Linguistically Fair? Quantisation Robustness across Bavarian, German, and English | Post-training quantisation reduces the memory cost of large language models, but it remains unclear whether compression preserves lang… | ↗Not briefed |
hn hn-ai | 29d | Show HN: Find paintings in any image's colors (HNSW from scratch, no server) | — | ↗Not briefed |
bluesky bluesky-ai | 29d | The people who think of their chatbot as a companion but want it switched off when they die is so morbidly hilarious. Sorry AI Smithers but… | The people who think of their chatbot as a companion but want it switched off when they die is so morbidly hilarious. Sorry AI Smither… | ↗Not briefed |
hn hn-ai | 29d | Show HN: Stream live open data in two commands | A service and open source tool to get open data streams in two commands. No sign up needed for subscribing.<p>Streams to try:<p>- <a h… | ↗Not briefed |
hn hn-ai | 29d | Show HN: Taskman – each ticket's Claude Code session, issue and PR in one TUI | — | ↗Not briefed |
hn hn-ai | 29d | Show HN: A coloring book where your drawings come alive and walks into a world | — | ↗Not briefed |
arxiv arxiv-ai | 29d | On the Optimality of Random Features under General Conditions: Misspecified Cases and Robust Learning | Random features provide a standard approach to scaling kernel methods to large datasets by approximating the kernel with a finite-dime… | ↗Not briefed |
hn hn-ai | 29d | Getty Images to File for Bankruptcy, Go Private in Restructuring Deal | — | ↗Not briefed |
hn hn-ai | 29d | Show HN: Sonia. Local Voice Assistant Linux. 13Gb RAM, No GPU, Skills Supported | — | ↗Not briefed |
rss techmeme | 29d | Apple unveils the fourth-gen Apple TV 4K, with Siri AI, an A19 chip, 128GB of storage, far-field mics, Wi-Fi 7, and Bluetooth 6, shipping November 13 for $299 (Jennifer Pattison Tuohy/The Verge) | Jennifer Pattison Tuohy / The Verge: Apple unveils the fourth-gen Apple TV 4K, with Siri AI, an A19 chip, 128GB of storage, far-field… | ↗Not briefed |
hn hn-ai | 29d | Port of the TypeScript compiler, checker and lsp to C# | — | ↗Not briefed |
arxiv arxiv-ai | 29d | Probing the Geometry of Color Representations in Multilingual Language Models | Do multilingual language models learn a universal internal geometry for color, and is it perceptually grounded or primarily linguistic… | ↗Not briefed |
arxiv arxiv-ai | 29d | AssetForge: Asset-Augmented Task Synthesis for Business Automation Agents | Training business automation agents requires executable tasks that connect a natural-language request to an application state and a ve… | ↗Not briefed |
bluesky bluesky-ai | 29d | 現状のLLMはそもそもの基盤データが無断データトレーニングが成り立っている技術なので、ローカルなら安心して使える／使えない以前の問題なんだよな…… 倫理的に使えません…… という認識です。 | 現状のLLMはそもそもの基盤データが無断データトレーニングが成り立っている技術なので、ローカルなら安心して使える／使えない以前の問題なんだよな…… 倫理的に使えません…… という認識です。 | ↗Not briefed |
bluesky bluesky-ai | 29d | Everything is LLM "written" in Rust by people who can barely describe what they've supposedly created because they were already depending o… | Everything is LLM "written" in Rust by people who can barely describe what they've supposedly created because they were already depend… | ↗Not briefed |
rss techmeme | 29d | Apple unveils the $149 HomePod mini 2, with an S10 chip, "Sound Glow" interactive lighting, and Siri AI via a future update, in five colors, shipping October 30 (Apple) | Apple: Apple unveils the $149 HomePod mini 2, with an S10 chip, “Sound Glow” interactive lighting, and Siri AI via a future update, in… | ↗Not briefed |
arxiv arxiv-ai | 29d | Jev-Guard: Fast Routing and Auditing Agent Actions over Editable Policy Trees | Agent harnesses require a low-latency guardrail model together with customizable audit policies. However, deploying an auto-regressive… | ↗Not briefed |
arxiv arxiv-ai | 29d | ShoeFit: A New Dataset and Dual-image-stream DiT Framework for Virtual Footwear Try-On | Virtual footwear try-on (VFTON), a critical yet underexplored area in virtual try-on (VTON), aims to synthesize faithful try-on result… | ↗Not briefed |
rss techmeme | 29d | Apple unveils the Apple Home Collection, a lineup of smart home accessories made with LG and Schneider, including cameras, thermostats, a doorbell, and a lock (Apple) | Apple: Apple unveils the Apple Home Collection, a lineup of smart home accessories made with LG and Schneider, including cameras, ther… | ↗Not briefed |
bluesky bluesky-ai | 29d | Mark Zuckerberg is a monster. | Mark Zuckerberg is a monster. | ↗Not briefed |
hn hn-ai | 29d | Show HN: Vaguely Right, a Jev-based daily game – submit a clue to hit the target🔧 Consensus | Each round you get a spectrum with a target on it (Calm ----X-- Stressful). Name something that sits at the target: a job interview, a… | ↗Not briefed |
arxiv arxiv-ai | 29d | Over-parameterization Is Neither Robust nor Fragile in Matrix Sensing: A Geometric Theory of Solution Robustness | Over-parameterization is ubiquitous in modern machine learning, yet its effect on robustness remains poorly understood. We study this… | ↗Not briefed |
rss techcrunch-ai | 29d | Amex introduces a framework to prevent AI agents from making unauthorized purchases | American Express is introducing a new framework designed to ensure AI agents only make purchases their users have authorized, potentia… | ↗Not briefed |
rss theverge-ai | 29d | Apple launches HomeView, its first smart display for the home | The HomeView is a smart speaker with a touchscreen display. It’s available in silver, starlight, rose pink, and space grey. | Image: A… | ↗Not briefed |
hn hn-ai | 29d | NewTube: A fast, touch-first YouTube client for Android | — | ↗Not briefed |
hn hn-ai | 29d | Show HN: Glossdex – semantic search that returns nothing when nothing matches | — | ↗Not briefed |
arxiv arxiv-ai | 29d | Complex materials engineering via Provenance Action Graphs: potential, structure and demonstration | This work discusses and tests the potential of reaction graphs for material synthesis, in which events consist of reaction steps (acti… | ↗Not briefed |
hn hn-ai | 29d | Better Portal – A macOS TUI for Higher-Bitrate PS5 Remote Play | — | ↗Not briefed |
rss techmeme | 29d | Apple unveils the $299 Apple HomeView, a 7" smart display with an A18 chip, tabletop and wall-mounted versions, and "HomeView software", shipping November 13 (Julian Chokkattu/Wired) | Julian Chokkattu / Wired: Apple unveils the $299 Apple HomeView, a 7" smart display with an A18 chip, tabletop and wall-mounted versio… | ↗Not briefed |
bluesky bluesky-ai | 29d | 【ご教示くださると嬉しいです】 最近、頻繁に「ローカルLLMなら機密情報が漏れることはない、安心して使える」という話が飛び交うのですが、 真偽のところはいかがなのでしょうか？ クライアントから預かった資料そのままは入れないが、調べものや叩き台に使えば大丈夫という話が本当なの？？… | 【ご教示くださると嬉しいです】 最近、頻繁に「ローカルLLMなら機密情報が漏れることはない、安心して使える」という話が飛び交うのですが、 真偽のところはいかがなのでしょうか？ クライアントから預かった資料そのままは入れないが、調べものや叩き台に使えば大丈夫という話が本… | ↗Not briefed |
arxiv arxiv-ai | 29d | Prompt or Policy? Optimizing Physician Slate Presentations with LLM User Simulators | Choosing the right physician is a key challenge for patients seeking medical care. Physician recommendation systems identify relevant… | ↗Not briefed |
arxiv arxiv-ai | 29d | SimInter: Evidence-Grounded User Simulation with Interview-Based Profile Enrichment for Conversational Recommender Systems | Conversational Recommender Systems (CRSs) require reliable user simulators for cost-effective evaluation. LLM-based simulators have ma… | ↗Not briefed |
bluesky bluesky-ai | 29d | openly (obviously, overtly) claude-made, which is inherently a yikes, would like to see a more traditional (read: not llm/claude) assessmen… | openly (obviously, overtly) claude-made, which is inherently a yikes, would like to see a more traditional (read: not llm/claude) asse… | ↗Not briefed |
hn hn-ai | 29d | After an 8 year pause venerable LAN Messenger is back in development | — | ↗Not briefed |
arxiv arxiv-ai | 29d | Bayesian Optimization of CVaR with Unknown Hyperparameters | We study Bayesian optimization of the Conditional Value-at-Risk (CVaR) of an expensive black-box function that depends on a decision a… | ↗Not briefed |
arxiv arxiv-ai | 29d | AffectWorld: Benchmarking Emotion-Driven Future World Prediction | The world encompasses both physical environments and human actors. Predicting its evolution requires understanding both physical laws… | ↗Not briefed |
hn hn-ai | 29d | Show HN: Sota-anchor – check what was shipped since cut-off before rebuilding it | — | ↗Not briefed |
arxiv arxiv-ai | 29d | Efficient Safety Guardrails through Latent Dual-Hypothesis Reasoning with On-Demand Trace Reconstruction | Guardrail approaches that rely on reasoning-based large language models (LLM) improve safety decisions, but come with substantial infe… | ↗Not briefed |
arxiv arxiv-ai | 29d | A dataset of formalised theorem statements from the Annals of Mathematics | Recent progress in autoformalisation of mathematics using AI has been impressive. Measuring this progress requires datasets of formal… | ↗Not briefed |
bluesky bluesky-ai | 29d | i dislike AI cuz the people behind it are weird rapist white supremacists who openly want to usher in a Claude-coded Third Impact where non… | i dislike AI cuz the people behind it are weird rapist white supremacists who openly want to usher in a Claude-coded Third Impact wher… | ↗Not briefed |
bluesky bluesky-ai | 29d | 運営レポートDay3（10/13）。Threadsは26→37人、noteは1日15本出して売上0円。いちばんスキが付いたのは、話題の記事ではなく「自分の正直な数字」の記事でした（PV12・スキ8）。無料部分で今日の数字を全部出しています note.com/yu_claudec… | 運営レポートDay3（10/13）。Threadsは26→37人、noteは1日15本出して売上0円。いちばんスキが付いたのは、話題の記事ではなく「自分の正直な数字」の記事でした（PV12・スキ8）。無料部分で今日の数字を全部出しています note.com/yu_cl… | ↗Not briefed |
hn hn-ai | 29d | Show HN: Rollport – Copy new iPhone photos to a folder over USB | Hi HN, I built Rollport, a cross-platform app (Windows, Mac and Linux) to import photos/videos from my iPhone to my computer, bec… | ↗Not briefed |
arxiv arxiv-ai | 29d | LegalRewardBench: Benchmarking Reward Models for Legal Preference Evaluation | Reward models are central to preference alignment for legal tasks, yet existing reward model benchmarks are still primarily centered o… | ↗Not briefed |
arxiv arxiv-ai | 29d | Output Discipline Sets the Floor: A Falsification-Guided, Model-Agnostic Agent Protocol for Auditable Hypothesis-Driven Decision Making | What binds large language models on auditable, hypothesis-driven decision tasks, where every candidate can be checked against the evid… | ↗Not briefed |
bluesky bluesky-ai | 29d | "If an expert prompts a result, they take up the mantle of responsibility I describe above [of explaining what you did, what is new vs. wel… | "If an expert prompts a result, they take up the mantle of responsibility I describe above [of explaining what you did, what is new vs… | ↗Not briefed |