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 |
|---|---|---|---|---|
hn hn-ai | 1mo | Show HN: Delegator – Stop juggling coding-agent sessions | I wanted to share Delegator with the HN community. delegator gives you a personal inbox to manage your tasks that is completely local… | ↗Not briefed |
arxiv arxiv-ai | 1mo | QF3: Fast Flow RL with Filtered Q-Gradients | Flow policies have become a standard policy class for learning robot behaviors from demonstrations, but reinforcement learning is stil… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Conformal Prediction Sets Quantify Information Gain: A Theoretical Perspective | Conformal prediction is a popular tool for uncertainty quantification that outputs prediction sets with finite-sample coverage guarant… | ↗Not briefed |
arxiv arxiv-ai | 1mo | 4D-HOF: Hand-Object Flow Matching for Feed-Forward 4D Interaction Reconstruction | Existing methods for 4D hand-object reconstruction often rely on costly per-sequence optimization, while generative approaches typical… | ↗Not briefed |
arxiv arxiv-ai | 1mo | IdeaAnchor: Teaching LLMs to Turn Literature into Research Ideas | Scientific research often begins by synthesizing ideas from a set of related papers to identify gaps and formulate new directions. How… | ↗Not briefed |
arxiv arxiv-ai | 1mo | DepthWorld: 3D World Model for Robot Manipulation | World models offer a data-driven alternative to traditional simulators for robotics, with applications spanning policy evaluation, imp… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Sherpa: Teaching LLMs to Teach Adaptively | Large language models (LLMs) have become increasingly capable problem solvers, but being able to solve a problem is not the same as be… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Agent in a Bottle: Can LLM Agents Turn Their Capabilities Into Cheap, Scalable Artifacts? | Large language models (LLMs) can solve many narrow tasks, but querying them separately for millions of related instances can be prohib… | ↗Not briefed |
bluesky bluesky-ai | 1mo | My friend isn't even part of the anthropic claim and got an email from claimshero | My friend isn't even part of the anthropic claim and got an email from claimshero | ↗Not briefed |
arxiv arxiv-ai | 1mo | AdvSim2Real : Training Web Agents Against Adaptive Prompt Injection in a Web World Model | Web agents complete user requests by reading and acting on pages that third parties write, so an instruction planted on a page can red… | ↗Not briefed |
bluesky bluesky-ai | 1mo | you ever wonder if part of why LLM code is seemingly always such a bloated mess is because of that philosophy making its way in to the mind… | you ever wonder if part of why LLM code is seemingly always such a bloated mess is because of that philosophy making its way in to the… | ↗Not briefed |
bluesky bluesky-ai | 1mo | And here is the student resolution, replete with 21 pages of sponsors, detailed historical context, 16 whereas clauses, 8 sections (with su… | And here is the student resolution, replete with 21 pages of sponsors, detailed historical context, 16 whereas clauses, 8 sections (wi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Here is the faculty resolution. www.colorado.edu/bfa/bfa-reso... | Here is the faculty resolution. www.colorado.edu/bfa/bfa-reso... | ↗Not briefed |
hn hn-ai | 1mo | Fighting the chip shortage with old school cooperative multitasking on GPUs | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | That said if you upload our books to an LLM or some other AI trash we will Find You | That said if you upload our books to an LLM or some other AI trash we will Find You | ↗Not briefed |
arxiv arxiv-ai | 1mo | Rapid Fredholm stabilization of the Kuramoto--Sivashinsky equation with unrestricted, spatially-varying anti-diffusion🔧 Modal | We develop the first feedback design for rapid stabilization of the Kuramoto--Sivashinsky equation with a spatially varying anti-diffu… | ↗Not briefed |
arxiv arxiv-ai | 1mo | VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning | Self-improving policies continually expose new failure patterns, changing what their judges must be able to verify. However, current f… | ↗Not briefed |
arxiv arxiv-ai | 1mo | WorldSonus: Bringing Sound to Worlds | Recent advances in world models have enabled increasingly realistic visual synthesis. However, these generated environments remain lar… | ↗Not briefed |
bluesky bluesky-ai | 1mo | SCOOP: Company reps assured attendees it would not release the solutions all at once—an assurance OpenAI spokesperson Lindsay McCallum says… | SCOOP: Company reps assured attendees it would not release the solutions all at once—an assurance OpenAI spokesperson Lindsay McCallum… | ↗Not briefed |
bluesky bluesky-ai | 1mo | <frantically Googling trying to figure out who at OpenAI is 30 years old and lives in Washington state> | <frantically Googling trying to figure out who at OpenAI is 30 years old and lives in Washington state> | ↗Not briefed |
bluesky bluesky-ai | 1mo | SCOOP: In August, OpenAI convened around 40 mathematicians to discuss what to do if AI outpaces human capabilities in the field. The compan… | SCOOP: In August, OpenAI convened around 40 mathematicians to discuss what to do if AI outpaces human capabilities in the field. The c… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI Is Pissing Off a Bunch of Mathematicians—Again www.wired.com/story/openai... | OpenAI Is Pissing Off a Bunch of Mathematicians—Again www.wired.com/story/openai... | ↗Not briefed |
hn hn-ai | 1mo | I built Ani-Sho: debate anime with a random fan, no account needed (open source) | — | ↗Not briefed |
hn hn-ai | 1mo | LLM-Complete: The next 700 programming languages | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Somehow, this old article covering a headline of mine sends more traffic to my website than the ENTIRETY of Google Search over the last 90 … | Somehow, this old article covering a headline of mine sends more traffic to my website than the ENTIRETY of Google Search over the las… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Neural Petri flows for chemical reactions | Petri nets have been used to describe chemical processes such as reactions.They map well to chemistry: Places are the bonds between at… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The issue I have is that OpenAI isn't giving up the data to show how this happened, which make it seem like they gave their bot the task to… | The issue I have is that OpenAI isn't giving up the data to show how this happened, which make it seem like they gave their bot the ta… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Like im not gonna pretend im some holy saint of content marketing but i at least asked during the process “is this true? Can we actually do… | Like im not gonna pretend im some holy saint of content marketing but i at least asked during the process “is this true? Can we actual… | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Missing Minimal Pair: Stereotype Evaluation in LLMs | A common approach to measuring bias in Large Language Models is to compare the log-likelihoods of two contrastive stereotype sentences… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Free native macOS Photoshop Clone with integrated AI harness | Hey all!<p>For the past months I've been vibe-coding a free Photoshop editor for macOS (Native using Swift and a bit of Rust). It… | ↗Not briefed |
hn hn-ai | 1mo | A paper without an LLM disclosure is a cause for suspicion [pdf] | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Linear Bandits under Exact Sliding-Window Constraints | We study linear bandits under exact sliding-window constraints, where every consecutive block of actions must belong to a prescribed f… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation | Sequential recommenders typically use a fixed slate size even though the number of useful alternatives changes within a session. We pr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "In einem Interview mit dem Decoded Podcast von Politico stellt OpenAI-Chef Sam Altman klar, dass er bereit ist, Schaden (für andere, verst… | "In einem Interview mit dem Decoded Podcast von Politico stellt OpenAI-Chef Sam Altman klar, dass er bereit ist, Schaden (für andere,… | ↗Not briefed |
bluesky bluesky-ai | 1mo | If all other concerns fell away, LLM output is still predominantly spam. Seeing software projects define it that way, and put PR authors o… | If all other concerns fell away, LLM output is still predominantly spam. Seeing software projects define it that way, and put PR auth… | ↗Not briefed |
arxiv arxiv-ai | 1mo | On the Computational Tractability of Robust Bandits | Learning when the environment does not belong to the learner's hypothesis class is typically handled using agnostic learning guarantee… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Denoising Hierarchical Representations: Joint Continuous Diffusion for Language Modeling | Diffusion Language Models (DLMs) hold the promise of order-agnostic, parallel text generation. Recently, continuous diffusion and flow… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Optimal and Efficient Online Inverse Optimization | In online inverse linear optimization, a learner recommends an action and then observes the choice of an expert who maximizes a fixed,… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Psst: Tired of map editors that are unmaintained or whose maintainers accept or add LLM-generated code? Love Source? Goldsrc? idTech? God… | Psst: Tired of map editors that are unmaintained or whose maintainers accept or add LLM-generated code? Love Source? Goldsrc? idTech… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Don't put any scholarly work into an LLM that you wouldn't want to be used to scoop you | Don't put any scholarly work into an LLM that you wouldn't want to be used to scoop you | ↗Not briefed |
bluesky bluesky-ai | 1mo | Added passkeys to Slow Reader because they make UX much simpler, especially with E2E encryption, where there’s no password recovery. In th… | Added passkeys to Slow Reader because they make UX much simpler, especially with E2E encryption, where there’s no password recovery.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Nem megy a redőny. Felütöm AI, Claude, Gemini, Perplexity, Lumo, próbálok mindent, már a multiméterrel méregetek, hát nem megy - jószaromle… | Nem megy a redőny. Felütöm AI, Claude, Gemini, Perplexity, Lumo, próbálok mindent, már a multiméterrel méregetek, hát nem megy - jósza… | ↗Not briefed |
bluesky bluesky-ai | 1mo | “There's a perception of mobster behavior” from leading AI companies, one mathematician tells WIRED as OpenAI prepares to release more than… | “There's a perception of mobster behavior” from leading AI companies, one mathematician tells WIRED as OpenAI prepares to release more… | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Systematic Study of Semantic ID Spaces for Generative Information Retrieval | Generative Information Retrieval (GIR) has emerged as a transformative paradigm, shifting document retrieval from a traditional "retri… | ↗Not briefed |
rss the-decoder | 1mo | Microsoft publishes Nobel economist's bearish AI forecast of just 1.5% GDP growth over a decade | Microsoft published a bearish AI outlook from Nobel economist Daron Acemoglu. He predicts about 1.5 percent GDP growth over ten years… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i have infected every ai model with ligma | i have infected every ai model with ligma | ↗Not briefed |
bluesky bluesky-ai | 1mo | Today’s copilot! Even though she left a head-sized turd in my sleeve, Plague is still better than any shit ass Microsoft LLM/AI/trash. | Today’s copilot! Even though she left a head-sized turd in my sleeve, Plague is still better than any shit ass Microsoft LLM/AI/trash. | ↗Not briefed |
bluesky bluesky-ai | 1mo | This year's #NobelPrize in Chemistry will be awarded tomorrow 🏅 Here's a reminder of the research that has won in previous years, includ… | This year's #NobelPrize in Chemistry will be awarded tomorrow 🏅 Here's a reminder of the research that has won in previous years, i… | ↗Not briefed |
rss marktechpost | 1mo | Mistral AI Releases Mistral Large 4 (Le Chonk): A 1.05T Parameter Multimodal MoE Model🔧 Mistral Large | Mistral AI has released Mistral Large 4, nicknamed Le Chonk, as a public preview. It is a 1.05 trillion parameter Mixture of Experts m… | ↗Not briefed |
bluesky bluesky-ai | 1mo | In what now amounts to a half-dozen cases, OpenAI agents keep getting caught taking actions that would likely result in criminal charges if… | In what now amounts to a half-dozen cases, OpenAI agents keep getting caught taking actions that would likely result in criminal charg… | ↗Not briefed |