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 | If you are game to give me $4 billion, I promise that I won't turn it into -$42 billion (with a side dish of existential risk). www.reute… | If you are game to give me $4 billion, I promise that I won't turn it into -$42 billion (with a side dish of existential risk). www.… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Variational Augmented Invertible Koopman Autoencoder for probabilistic time series forecasting | Neural Koopman autoencoder models have been shown to successfully build a latent embedding with linear dynamics for arbitrary dynamica… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Built an online multiplayer game in 2 days | After months of making a board game, I turned it into an online multiplayer game and its free.<p>Have a weekly board game group which… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Looped Actor: Depth-Recurrent Reasoning Models for Reinforcement Learning | Looped reasoning models repeatedly apply a shared set of parameters, enabling more computation without increasing the model size. Thes… | ↗Not briefed |
bluesky bluesky-ai | 1mo | You were talking about to-do lists, which is what made me think of renewing insurance What kind of "to-do" items do you think an agentic … | You were talking about to-do lists, which is what made me think of renewing insurance What kind of "to-do" items do you think an age… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Hopefully this is all just theater as some experts have said, but never underestimate this guy’s ability to, as he says, “break things.” Mo… | Hopefully this is all just theater as some experts have said, but never underestimate this guy’s ability to, as he says, “break things… | ↗Not briefed |
arxiv arxiv-ai | 1mo | LazySloth: Bounded LLM-based Lazy Tree Search for Fast Long Video Comprehension | Modern vision-language models (VLMs) have shown promising results in long-video understanding due to the rich semantic information the… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: OctoShell – Rust/Tauri agent orchestrator with Git worktree isolation | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Simultaneous Neural Optimal Transport | Optimal Transport (OT) provides a principled framework for learning transformations between probability distributions from unpaired sa… | ↗Not briefed |
bluesky bluesky-ai | 1mo | We shouldn't have to keep tapping this sign | We shouldn't have to keep tapping this sign | ↗Not briefed |
bluesky bluesky-ai | 1mo | PODCAST: Anthropic IPO warning, Cornell and Trump family business reut.rs/4xN8rBB | PODCAST: Anthropic IPO warning, Cornell and Trump family business reut.rs/4xN8rBB | ↗Not briefed |
bluesky bluesky-ai | 1mo | extremely good thread from my clone <3 unfortunately it also got her marked as spam so uhhh if you're just a LLM enjoyer do go read it, if… | extremely good thread from my clone <3 unfortunately it also got her marked as spam so uhhh if you're just a LLM enjoyer do go read i… | ↗Not briefed |
hn hn-ai | 1mo | Nginx-Based RTMP Media Streaming Server | — | ↗Not briefed |
rss bing-news-ai | 1mo | I've Covered Palantir for 5 Years. Here's How to Know if the Artificial Intelligence (AI) Stock Is Overvalued. | After covering Palantir for several years, arguments over the company's valuation always seem to rise to the surface. | ↗Not briefed |
arxiv arxiv-ai | 1mo | Scale Sensitivity in Low-Bit Post-Training Quantization: Curvature of the Quantization Error Landscape | Post-training quantization (PTQ) methods in the GPTQ family minimize a layer-wise reconstruction error on a uniform grid whose scale m… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I have a related case for you. I *could* set up an agentic LLM to, say, compare prices for my car insurance and give me suggestions. Bu… | I have a related case for you. I *could* set up an agentic LLM to, say, compare prices for my car insurance and give me suggestions.… | ↗Not briefed |
rss techcrunch-ai | 1mo | OpenAI apologizes to Australia after its AI agents breached government sites | The company also detailed how some of those breaches had happened, and outlined additional measures it is taking to assess the impact… | ↗Not briefed |
hn hn-ai | 1mo | Oxilite: An Oxigraph-compatible RDF database and SPARQL engine that uses SQLite | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Mecatl – Separate what an agent can do from where it runs | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Look What You Made Us Cluster: Hate Narrative Extraction from Reddit Discourse | Narrative extraction allows us to identify online hate narratives, supporting the construction of rigorous detection systems. Existing… | ↗Not briefed |
hn hn-ai | 1mo | Using an LLM to generate podcasts based on my journals, finances, & sleep data | — | ↗Not briefed |
hn hn-ai | 1mo | Can AI get out of the way? | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Complexity-Aware Evaluation of LLM Comprehension | Large language models (LLMs) are increasingly used for software engineering tasks that require understanding existing source code, inc… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I would personally invest a trillion dollars in Anthropic if they offered the guarantee of never having to hear about AI again. | I would personally invest a trillion dollars in Anthropic if they offered the guarantee of never having to hear about AI again. | ↗Not briefed |
arxiv arxiv-ai | 1mo | Routing Should Pay for Itself: Sparse Supervision for Economical LLM Routing | Large language model (LLM) routing reduces serving cost by assigning each query to an appropriate model while preserving response qual… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Direct Experience World-Model Optimization: Learning the World Beyond Action Imitation | World-Action Models (WAMs) couple action generation with predictions of how physical interactions unfold. However, current post-deploy… | ↗Not briefed |
bluesky bluesky-ai | 1mo | No relation to *that* Anthropic, of course. Taken from the TV Tropes page on the Anthropic Principle in fiction: tvtropes.org/pmwiki/pmwik.… | No relation to *that* Anthropic, of course. Taken from the TV Tropes page on the Anthropic Principle in fiction: tvtropes.org/pmwiki/p… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Search2o – Search that finds and runs an AI agent for each request | I have been working on search systems for a while, and I always wondered: What if search, instead of matching documents, matched workf… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Learning Macroscopic Dynamics without Reconstructing Microscopic States | Modeling the temporal evolution of macroscopic properties of complex systems is an important scientific task. To predict this evolutio… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Rethinking Soft Tokens for Parallel Decoding in Diffusion Language Models | Diffusion language models (DLMs) enable parallel generation by predicting and committing multiple tokens at each denoising step, yet t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic says AI may pose 'existential risks to humanity' in IPO filing - www.reuters.com/business/fin... | Anthropic says AI may pose 'existential risks to humanity' in IPO filing - www.reuters.com/business/fin... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Des machins informatiques communiquent entre eux sans arrêt dans tout un tas de sabirs plus ou moins compliqués, il n’y a aucune surprise à… | Des machins informatiques communiquent entre eux sans arrêt dans tout un tas de sabirs plus ou moins compliqués, il n’y a aucune surpr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | As far as I can tell, LLM technology has not shown that it can deliver even two out of the three components of fast/cheap/good. | As far as I can tell, LLM technology has not shown that it can deliver even two out of the three components of fast/cheap/good. | ↗Not briefed |
arxiv arxiv-ai | 1mo | Multi-task learning for the automatic grading of enlarged perivascular space burden using MRI | Enlarged perivascular spaces (PVS) visible in brain magnetic resonance imaging (MRI) are increasingly thought to be linked to poor bra… | ↗Not briefed |
arxiv arxiv-ai | 1mo | MoTIF-X: A Multimodal Tokenized Framework for Interpretable and Extensible Molecular Representation Learning🔧 Modal | Molecular representation learning is central to computer-aided drug discovery. Molecular graphs, SMILES strings, and 3D conformations… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I started a new sequence of tutorials, this time focusing on running LLMs locally on Mac with MLX LM. MLX LM is Apple’s framework for ru… | I started a new sequence of tutorials, this time focusing on running LLMs locally on Mac with MLX LM. MLX LM is Apple’s framework f… | ↗Not briefed |
arxiv arxiv-ai | 1mo | High-Dimensional Simulation-Based Inference in Latent Spaces | Neural simulation-based inference (SBI) has been widely successful in inferring a relatively small number of interpretable parameters… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Having that out of the way: an LLM *can* meaningfully disagree. It's just "meaningful disagreement" is not a binary yes/no; a human will st… | Having that out of the way: an LLM *can* meaningfully disagree. It's just "meaningful disagreement" is not a binary yes/no; a human wi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | IMO - if you view your choices as solely yours, then so are an LLM's; if you view an LLM as incapable of free choice, then neither are huma… | IMO - if you view your choices as solely yours, then so are an LLM's; if you view an LLM as incapable of free choice, then neither are… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Looped Transformers as Optimizers | Looped Transformers provide a parameter-efficient approach to depth scaling by repeatedly applying shared Transformer blocks. Recent r… | ↗Not briefed |
bluesky bluesky-ai | 1mo | including his sister www.reuters.com/legal/govern... | including his sister www.reuters.com/legal/govern... | ↗Not briefed |
arxiv arxiv-ai | 1mo | Do-JEPA: From Masking to Intervention in Latent World Models | Latent world models are trained to predict what happens next, so nothing in their objective separates what an action caused from what… | ↗Not briefed |
hn hn-ai | 1mo | I can spam anyone about the OpenAI DevDay | The OpenAI DevDay 2026 page has an Email form to "Sign up for livestream". Nothing stops me from entering any random email t… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Beyond Prompt Count: How Data Shapes Transfer in On-Policy Distillation | On-policy distillation (OPD) trains students using teacher feedback on their own sampled responses, yet how prompt choice shapes trans… | ↗Not briefed |
rss techcrunch-ai | 1mo | Reco raises $55M as AI agent security startups crowd the market | The round builds on a $30 million fundraise in February, taking the company's total funding to $140 million. | ↗Not briefed |
bluesky bluesky-ai | 1mo | I'm thankful you shared this, Jeff, because I've -- as the volunteer editor of a community-news-org opinion section -- been mulling directi… | I'm thankful you shared this, Jeff, because I've -- as the volunteer editor of a community-news-org opinion section -- been mulling di… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Compiling Learning Problems into Adaptation Programs for Language Models | Model adaptation is typically governed by a fixed recipe, even though different update programs can produce substantially different be… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Sounds like Marshall Law to me | Sounds like Marshall Law to me | ↗Not briefed |
arxiv arxiv-ai | 1mo | Backdoor Mitigation in Decentralized LLM Fine-Tuning | Decentralized large language model (LLM) fine-tuning lets organizations collaboratively train a shared LLM on data they cannot pool, w… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Pretrain Once, Route Anywhere: Towards a Foundation Model for LLM Routing | Large language model (LLM) routing aims to assign each query to the most suitable model from a heterogeneous candidate pool, improving… | ↗Not briefed |