News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

24h 48h 7d 30d 90d All (6mo)

Source	Age	Title	Summary	Status
arxiv arxiv-ai	20d	PaceVLA: Adaptive Temporal Action Quantization in Vision-Language-Action Models	Fast and reliable manipulation is crucial for practical robot deployment. Although Vision-Language-Action (VLA) models have recently s…	↗Not briefed
arxiv arxiv-ai	20d	EmbodiedWorld-200K: Toward Embodied World Agents via Embodied-Planning Dataset and Interactive World Models	Building embodied agents that can perceive, reason, and act across diverse 3D worlds is a central research direction toward artificial…	↗Not briefed
arxiv arxiv-ai	20d	Productive State-Space Exploration for Multi-turn RL in LLM Agents	Reinforcement Learning (RL) for Large Language Model (LLM) agents in multi-turn, long-horizon environments remains challenging due to…	↗Not briefed
bluesky bluesky-ai	20d	oh no i owe anthropic one cent theyre gonna send the claude legbreakers after me	oh no i owe anthropic one cent theyre gonna send the claude legbreakers after me	↗Not briefed
arxiv arxiv-ai	20d	CXBench: A Benchmark for Process-Compliant LLM Agents in Regulated Customer-Experience Workflows	Tool-using language agents increasingly perform regulated customer-service workflows, where a correct final state is insufficient: age…	↗Not briefed
bluesky bluesky-ai	20d	Machtmissbrauch von #Meta KI Zugriffszwang für Medien Webseiten, wenn sie #Facebook #Instagram #Whatsapp & Co. nutzen! #DeleteMeta #FckZc…	Machtmissbrauch von #Meta KI Zugriffszwang für Medien Webseiten, wenn sie #Facebook #Instagram #Whatsapp & Co. nutzen! #DeleteMeta #…	↗Not briefed
rss techmeme	20d	A profile of Alona Katz, chief of Brooklyn DA's Virtual Currency Unit, who recently led an operation that seized and disrupted over 240 crypto fraud websites (Bloomberg)	Bloomberg: A profile of Alona Katz, chief of Brooklyn DA's Virtual Currency Unit, who recently led an operation that seized and disrup…	↗Not briefed
arxiv arxiv-ai	20d	AdaExploreBench: Benchmarking LLM Agents for Proactive Exploration and Test-time Adaptation in Evolving Environments	Large Language Model (LLM) agents are increasingly deployed in unfamiliar and evolving environments, requiring them to proactively exp…	↗Not briefed
arxiv arxiv-ai	20d	OrderGPT: Generation Order as a Learnable Modeling Choice	Generative modeling of discrete data has two popular paradigms: autoregressive models such as GPT, which factorize a sequence in a lef…	↗Not briefed
arxiv arxiv-ai	20d	Claim Laundering: When Agent Decision Models Trust Records Over Reality	Agent decision models select actions in a single forward pass from inputs that combine environmental observations with the agent's own…	↗Not briefed
arxiv arxiv-ai	20d	Policy Search in Code Space for Multi-Agent Pathfinding	Developing coordination policies for complex multi-agent systems relies on either labor-intensive algorithm design or opaque task-spec…	↗Not briefed
arxiv arxiv-ai	20d	Speech LLMs That Breathe: Modeling Non-Verbal Vocalizations in Dialogue	The user experience of speech interaction models has improved, benefiting from the modeling of paralinguistic attributes and advances…	↗Not briefed
arxiv arxiv-ai	20d	Global probabilistic forecasting of air pollution	Fine particulate matter (PM2.5) refers to particulate matter with an aerodynamic diameter of 2.5 micrometre or less. It comes from hum…	↗Not briefed
arxiv arxiv-ai	20d	MITRA-lexicon: Lexica of attested translations from Sanskrit into English, Tibetan and Chinese	The standard Sanskrit lexica are monumental and indispensable, but they cite their attestations as abbreviated references to editions,…	↗Not briefed
arxiv arxiv-ai	20d	The KV-Surgeon: Compensation-Aware Offline KV-Cache Compaction	Offline compaction of the key--value cache (KV-cache) reduces the cost of repeatedly querying a fixed context by constructing a smalle…	↗Not briefed
bluesky bluesky-ai	20d	Like, LLM can only do something competent when there's a person who actually know the craft behind. The ai "art" that tricks people was pr…	Like, LLM can only do something competent when there's a person who actually know the craft behind. The ai "art" that tricks people w…	↗Not briefed
arxiv arxiv-ai	20d	Keep the Lineages Alive: Diversity-Preserving SMC for Diffusion LLM Decoding	Inference-time sequential Monte Carlo (SMC) decoding can improve the reasoning accuracy of masked diffusion language models (diffusion…	↗Not briefed
arxiv arxiv-ai	20d	EvoSkill v2: Fully Agentic Skill Evolution from Benchmark Adaptation to a Stateful Evolver	EvoSkill v2 is a three-agent system that edits the instructions (skills) an agent is given, leaving its model and harness intact. The…	↗Not briefed
hn hn-ai	20d	Ask HN: How do you cap what your coding agents can spend?	I run autonomous coding agents and I'm trying to understand how other people handle money when an agent needs to pay for things (…	↗Not briefed
arxiv arxiv-ai	20d	TESS: LLM-guided Translation, Exploration, Simulation, and Search for Embodied Task Planning	In this work we focus on the problem of embodied task planning, which requires an agent to accomplish a complex task by separating it…	↗Not briefed
bluesky bluesky-ai	20d	Anthropic says around 20% of the company’s shares will be dedicated to “the public good” when the new fund is combined with existing pledge…	Anthropic says around 20% of the company’s shares will be dedicated to “the public good” when the new fund is combined with existing p…	↗Not briefed
arxiv arxiv-ai	20d	UtilityBench: Measuring Language Model Utility When Errors Have Different Costs🔧 Elicit	As language models are increasingly deployed in real-world scenarios with important outcomes, the impact of their mistakes often exten…	↗Not briefed
bluesky bluesky-ai	20d	知識や経験のない人がLLMを使っても大して得るものはないだろう。ニーチェを読んでいない人がLLMでニーチェの話をしても意味がない。私は高等数学をよく知らないので高度な話はできないことを理解しつつ、高校数学をやり直している。LLMは底上げをするのであって、0を1にはしない。	知識や経験のない人がLLMを使っても大して得るものはないだろう。ニーチェを読んでいない人がLLMでニーチェの話をしても意味がない。私は高等数学をよく知らないので高度な話はできないことを理解しつつ、高校数学をやり直している。LLMは底上げをするのであって、0を1にはしな…	↗Not briefed
hn hn-ai	20d	Ask HN: Do you also feel that it's impossible to keep up with AI?	I have a tech background, I haven't been super active in coding or other things, but I still play with stuff a lot. And yet, in t…	↗Not briefed
arxiv arxiv-ai	20d	Maestro: Agentic Formal Proof Development by Coordination	Lean is increasingly used to verify mathematical proofs and software produced by LLM agents, but writing formal proofs remains hard. E…	↗Not briefed
arxiv arxiv-ai	20d	LEAPS: Long-context Evaluation of Adaptive Pattern Learning and State Tracking	Long-context language models (LCLMs) power agents that act over long trajectories, from improving a codebase to running research exper…	↗Not briefed
arxiv arxiv-ai	20d	Secure Agentic Coding through Counterexample-Grounded Feedback	As coding models and agent harnesses become more capable, reliable security verification is becoming a central bottleneck. Improving s…	↗Not briefed
bluesky bluesky-ai	20d	こんにちはっ✨ リプありがとう🫶 #AIイラスト #ATelier #ATelierで生成	こんにちはっ✨ リプありがとう🫶 #AIイラスト #ATelier #ATelierで生成	↗Not briefed
arxiv arxiv-ai	20d	FisherEdit: Fisher Information Guided Knowledge Editing for Large Language Models	Sequential knowledge editing in Large Language Models (LLMs) often leads to a degradation of general capabilities due to cumulative pa…	↗Not briefed
bluesky bluesky-ai	20d	Well, it took a law firm stealing from another law firm to finally make someone in power angry, but I hope this is the precedent we need to…	Well, it took a law firm stealing from another law firm to finally make someone in power angry, but I hope this is the precedent we ne…	↗Not briefed
bluesky bluesky-ai	20d	生成AIってなんつかwめちゃくちゃ難しいんですよwいやLLMってむずかしくない？だから「なんでお話してる生成AIとプログラミングの生成AIと絵が出る生成AIと Xと！chatGTPを！一緒に語るんですか！」ってなるんだと思うんだけど他分野の権利侵害に目を当てようって…	生成AIってなんつかwめちゃくちゃ難しいんですよwいやLLMってむずかしくない？だから「なんでお話してる生成AIとプログラミングの生成AIと絵が出る生成AIと Xと！chatGTPを！一緒に語るんですか！」ってなるんだと思うんだけど他分野の権利侵害に目を当…	↗Not briefed
arxiv arxiv-ai	20d	CALM: Consistency-Aware Logit Masking in Decoding Space for Mitigating Overthinking of Large Reasoning Models	Large reasoning models (LRMs) have achieved strong performance on challenging mathematical and scientific tasks through long chain-of-…	↗Not briefed
arxiv arxiv-ai	20d	An Analysis of Semantic and Spatio-Temporal Co-evolution for Edge Classification on Dynamic Text-Attributed Graphs	Dynamic edge classification on text-attributed Web graphs, such as review networks, community forums, and news-derived knowledge graph…	↗Not briefed
hn hn-ai	20d	iOS running on the Redmi Note 10 Pro	—	↗Not briefed
hn hn-ai	20d	Show HN: Sauva – Terminal Unicode Explorer	I made a tool to search and inspect Unicode characters in the terminal.<p>- Browse Unicode blocks or search by character, code point,…	↗Not briefed
arxiv arxiv-ai	20d	Out-of-Distribution Detection in Continual Learning	Out-of-distribution detection in continual learning (OOD-CL) is highly challenging, as it requires (i) learning a sequence of tasks wi…	↗Not briefed
bluesky bluesky-ai	20d	x.com/blackright0/... ※※Xリンク注意※※ Googleがgemini入りのスマホCMをバンバン流し、TVのワイドショーでは有識者らしき人が「パスワードの使い回しはやめよう、大手のAIなら情報漏洩は起きないから怪しいAIさえ使わなければ大丈夫」とか視聴者…	x.com/blackright0/... ※※Xリンク注意※※ Googleがgemini入りのスマホCMをバンバン流し、TVのワイドショーでは有識者らしき人が「パスワードの使い回しはやめよう、大手のAIなら情報漏洩は起きないから怪しいAIさえ使わなければ大丈夫」…	↗Not briefed
arxiv arxiv-ai	20d	SWE-Residual: Evaluating Starting-State Robustness of Coding Agents	Software engineering (SWE) benchmarks are the primary way to measure the capabilities of coding agents: an agent is placed in a code r…	↗Not briefed
hn hn-ai	20d	FreeLeaf: An free and open latex collaborative suite	—	↗Not briefed
arxiv arxiv-ai	20d	Budgeted Attack Scheduling for LLM Red-Teaming: The Roles of Static Priors, Semantic Guidance, and Online Feedback🔧 GPT-5	As jailbreak attacks against large language models (LLMs) become increasingly diverse, red-teaming has come to rely on large libraries…	↗Not briefed
arxiv arxiv-ai	20d	When Should AI Chime In? Benchmarking Timely Contribution of Proactive Agents in Multi-Party Chat	Proactive agents can support group collaboration by surfacing useful information without waiting to be asked, but even relevant contri…	↗Not briefed
bluesky bluesky-ai	20d	Guuuuys, guuus, mister Yann "Gears" " post-LLM" LeCun is ripping out Elon's rectum on hellsite dot birb	Guuuuys, guuus, mister Yann "Gears" " post-LLM" LeCun is ripping out Elon's rectum on hellsite dot birb	↗Not briefed
arxiv arxiv-ai	20d	Motor compound fault diagnosis using an attention-based cross-talk architecture	As physical AI increasingly relies on sensing to understand physical systems, reliable condition monitoring of industrial machinery ha…	↗Not briefed
hn hn-ai	20d	GeoParquet Downloader for QGIS	—	↗Not briefed
arxiv arxiv-ai	20d	Manifold-Constrained Residual Adaptation with Frame-Adaptive Layer Aggregation for Speaker Verification	Large self-supervised speech models provide powerful representations for speaker verification, but full fine-tuning is expensive. Most…	↗Not briefed
rss bing-news-ai	20d	This Artificial Intelligence (AI) Stock Is Already Seeing Demand Stretching Into 2030	ASML recently reported its earnings, highlighting strong demand in the coming years.	↗Not briefed
rss bing-news-ai	20d	This artificial intelligence (AI) stock is already seeing demand stretching into 2030	ASML recently reported its earnings, highlighting strong demand in the coming years.	↗Not briefed
rss bing-news-ai	20d	This artificial intelligence (AI) stock is already seeing demand stretching into 2030	ASML recently reported its earnings, highlighting strong demand in the coming years.	↗Not briefed
rss bing-news-ai	20d	This artificial intelligence (AI) stock is already seeing demand stretching into 2030	ASML recently reported its earnings, highlighting strong demand in the coming years.	↗Not briefed
arxiv arxiv-ai	20d	Learn the Interface, Not the Implementation: Output Representation Dominates Training Regime in LLM-Based RTL Generation	Large-language-model register-transfer-level (RTL) generation has concentrated on how to train the model: fine-tuning, continued pre-t…	↗Not briefed