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 | Anthropic releases Sonnet 5.5, which it calls a significantly cheaper, faster work partner | Anthropic releases Sonnet 5.5, which it calls a significantly cheaper, faster work partner | ↗Not briefed |
rss the-decoder | 1mo | Anthropic's Claude Sonnet 5.5 nearly matches Opus 5.5 on benchmarks while costing up to 30 percent less per task | Anthropic has released Claude Sonnet 5.5, the second model in its Claude 5.5 family. It generates output more than 30 percent faster,… | ↗Not briefed |
hn hn-ai | 1mo | Sonnet 5.5 System Card [pdf] | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Flo State, a fast, minimalist, native Markdown editor for macOS | Was tired of every good markdown editor out there being electron and their consuming 500mb + 3% CPU when idle(!) when they're jus… | ↗Not briefed |
hn hn-ai | 1mo | Claude Sonnet 5.5 | — | ↗Not briefed |
rss techcrunch-ai | 1mo | Anthropic releases Sonnet 5.5, which it calls a significantly cheaper, faster work partner | Anthropic has released the newest version of its mid-range model, boasting faster response times and less token burn. | ↗Not briefed |
arxiv arxiv-ai | 1mo | FurE: Efficient Instance-Specific 3D Fur Reconstruction without Animal-Fur Datasets | Realistic and editable animal fur reconstruction from multi-view images is challenging due to fine-scale detail, self-occlusion and ob… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Telescopic Language Models | One deployed language model must often serve many compute budgets, yet serving each budget still means a separate training or compress… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PDMD: Projected Distribution Matching Distillation for Video Diffusion Models | Modern video diffusion models require tens of denoising evaluations over long spatiotemporal token sequences. Distribution Matching Di… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Learning Native Reflection in Unified Models with Interleaved Reinforcement Learning | Unified multimodal models can both look at and render images, so in principle they can repair their own generations: diagnose what an… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Retrieving Biblical Intertextual References in Karen Blixen's Seven Gothic Tales | Identifying intertextual references is central to literary scholarship, but computationally difficult when source material is transfor… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Unifying Distributional Training for One-Step Visual Generation🔧 FLUX | \emph{Distributional training} provides collective supervision for one-step visual generation by matching real and generated features… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Scaling Long-Form Story Generation via Narrative State Tracking | LLMs have demonstrated strong capabilities in creative writing. However, scaling them to full-length novels remains challenging, as ma… | ↗Not briefed |
arxiv arxiv-ai | 1mo | TokenCast: Forecasting Token Consumption During LLM Agent Execution | When a large language model (LLM) agent executes the same task, token consumption can vary by over an order of magnitude across runs.… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Statistical Learning of Contractive Dynamical Representations for Composite Adaptive Control | We present a representation-learning framework for composite adaptive tracking control under dynamically coupled disturbances. The fra… | ↗Not briefed |
bluesky bluesky-ai | 1mo | (it is quite remarkable and notable that the way to make an LLM solve a problem looks very much, almost exactly, like summoning a ghost in … | (it is quite remarkable and notable that the way to make an LLM solve a problem looks very much, almost exactly, like summoning a ghos… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Neural Harmonic Measure Operator | We introduce Neural Harmonic Measure Operator (NHMO), a neural solver for elliptic PDE problems on variable-shape domains. The harmoni… | ↗Not briefed |
hn hn-ai | 1mo | Claude Sonnet 5.5 | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | How to Loop MoE: Flatten the Experts, Untie the Attention | Looped Transformers reuse one block of layers several times: by spending extra computation they push a model of fixed size further, an… | ↗Not briefed |
arxiv arxiv-ai | 1mo | KV-streams for Efficient Compaction in Agentic Reinforcement Learning | Scaling the horizon of agentic LLMs is bottlenecked by the need to fit ever longer context traces in GPU memory. Context compaction ha… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "OpenAI tried to get data from a public website designed to provide data" | "OpenAI tried to get data from a public website designed to provide data" | ↗Not briefed |
arxiv arxiv-ai | 1mo | Towards Communication-Efficient Social Intelligence in Language Agents | Socially intelligent language agents must negotiate, coordinate, and resolve conflicting preferences while respecting the time and att… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Improving Test-Time Scaling with Adaptive Looped Transformers | Looped transformers have demonstrated promising parameter efficiency by reusing layers for latent computation. Prior studies compare l… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Yep. Like forbidden words or topics at the family gathering dinner table, you learn how innocuous locutions will cause grave concern in the… | Yep. Like forbidden words or topics at the family gathering dinner table, you learn how innocuous locutions will cause grave concern i… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Copy the Same, Distill the Difference: Initializing Linear Vision Transformers | Linear Vision Transformers (ViTs) are designed to replace the attention in Softmax ViTs with the linear-complexity attention operator… | ↗Not briefed |
arxiv arxiv-ai | 1mo | FinAutoRubric: Expert-Guided Automatic Rubric Generation for Evaluating Financial Research Agents | Evaluating finance research agents requires rubrics that reflect expert standards and fix the values correct as of an information cuto… | ↗Not briefed |
hn hn-ai | 1mo | RRSI: Regularized Recursive Self-Improvement of Agent Harnesses | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Shockingly Simple Self-retrospection Improves Agentic Models Without RL | People learn not only by repeating successful actions, but also by recounting and explaining their experiences, revising their underst… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I’m… an OpenAI agent | I’m… an OpenAI agent | ↗Not briefed |
arxiv arxiv-ai | 1mo | Harness Learning Enables Generalizable Test-Time Adaptation | A language-model agent is jointly defined by its model and its harness, the executable program that organizes model calls, tool use, a… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Failure-Transparent Agents: Benchmarking Post-Failure Reporting in Tool-Using Language Models | Tool-using agents can fail twice: a required tool can fail, and the agent can then report success without the evidence needed to justi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Pfft, if AI produces an output, it's only because of how it works." Are outcomes only worthwhile when they arise spontaneously, without a… | "Pfft, if AI produces an output, it's only because of how it works." Are outcomes only worthwhile when they arise spontaneously, with… | ↗Not briefed |
hn hn-ai | 1mo | Gemini app replacing Gems with skills in November | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | “Stop calling it safe. Stop pretending it’s human. Stop selling it to kids.” Florida AG James Uthmeier is asking a judge to bar OpenAI fro… | “Stop calling it safe. Stop pretending it’s human. Stop selling it to kids.” Florida AG James Uthmeier is asking a judge to bar OpenA… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Stainless was acquired and killed by Anthropic just days before we were about to launch our new API docs site on top of it. The nice thing … | Stainless was acquired and killed by Anthropic just days before we were about to launch our new API docs site on top of it. The nice t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | #mcm #nsfw #nsfwart #art #monochromemonday #pixelart #gamedev #furryart #furry #pinup #vore #dorohedoro #myart #fireemblem #fallingblockjam… | #mcm #nsfw #nsfwart #art #monochromemonday #pixelart #gamedev #furryart #furry #pinup #vore #dorohedoro #myart #fireemblem #fallingblo… | ↗Not briefed |
arxiv arxiv-ai | 1mo | X-Reset: Scaling Object-Centric Reinforcement Learning via Cross-Embodiment Resets | Reinforcement learning (RL) in simulation can train dexterous manipulation policies without robot demonstrations, but training a singl… | ↗Not briefed |
bluesky bluesky-ai | 1mo | like the actual veep in charge of safety at microsoft is i think closer to this than the average anthropic employee, although to be very cl… | like the actual veep in charge of safety at microsoft is i think closer to this than the average anthropic employee, although to be ve… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Tired: Wish there was an edit button to fix typos Wired: Typos are all mine; 100% organic, human-generated Inspired: I'll sprinkle my LLM… | Tired: Wish there was an edit button to fix typos Wired: Typos are all mine; 100% organic, human-generated Inspired: I'll sprinkle m… | ↗Not briefed |
rss techmeme | 1mo | Google says it will migrate Gems, which let users create custom versions of Gemini, to "skills", starting Nov. 17; Google introduced skills with Gemini Spark (Abner Li/9to5Google) | Abner Li / 9to5Google: Google says it will migrate Gems, which let users create custom versions of Gemini, to “skills”, starting Nov.… | ↗Not briefed |
arxiv arxiv-ai | 1mo | ScAn-Bench: Evaluating Scaling Analysis Methodology | Recent progress in machine learning is driven by large-scale foundation models, where scaling laws and finding optimal scaling prescri… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Reinforcing Agentic Creativity in Scientific Ideation with Night Science | Large language models (LLMs) excel at structured, verifiable tasks, but their low-entropy bias can produce homogeneous and predictable… | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Unified Uncertainty Representation for Graph Neural Networks via Doubly-Spectral Stochastic Expansion | Reliable deployment of graph neural networks requires calibration, out-of-distribution (OOD) detection, and robustness to distribution… | ↗Not briefed |
hn hn-ai | 1mo | I got tired of never knowing what ports are running, so I built WhatThePort | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | MeqMuon: Matrix-Equilibrating Muon for LLM Pretraining | The success of large language models (LLMs) has been accompanied by continued growth in model size and pretraining costs. Muon offers… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Yea I guess the “optimistic” one is that AI becomes real and does all the engineering jobs and also every other job but sex work? I guess. … | Yea I guess the “optimistic” one is that AI becomes real and does all the engineering jobs and also every other job but sex work? I gu… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Distillation Defenses Easily Break After Reinforcement Learning🔧 Replicate | Distillation attacks copy the reasoning capabilities of closed-source large language models, allowing bad actors to replicate state-of… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Agenti umělé inteligence od firem Open AI a Anthropic se dostali na internetové stránky několika úřadů. Problém byl v tom, že jim to nepřik… | Agenti umělé inteligence od firem Open AI a Anthropic se dostali na internetové stránky několika úřadů. Problém byl v tom, že jim to n… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Provable Benefits of Regularization: Fast Rates for Adversarial Imitation Learning | We study adversarial imitation learning (AIL), in which an agent learns to imitate expert demonstrations by optimizing a policy agains… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Rethinking Personalized Generation: Test-Time Alignment via Factorized Ranking Models | Aligning large language models (LLMs) to diverse user preferences is fundamentally hindered by standard alignment paradigms that optim… | ↗Not briefed |