All published intelligence, sovereign stack research, ecosystem news, and AI foundation model reviews.
arXiv:2610.20764v1 Announce Type: new Abstract: This paper challenges the understanding capabilities of existing AI paradigms and emerging world models by asking a fundamental question: How quickly can reality change before an intelligent system can no longer reliably perceive,…
arXiv:2610.20676v1 Announce Type: new Abstract: Correctness feedback suppresses the very evidence that specification-completeness auditing is designed to uncover: programs that satisfy a written contract but violate intended behavior. We demonstrate this objective reversal on 12…
arXiv:2610.20822v1 Announce Type: new Abstract: Text-to-video models often generate the visual concepts requested in natural language prompts but fail to satisfy their intended ordering, timing, and persistence requirements. We propose representing these requirements as…
arXiv:2610.20667v1 Announce Type: new Abstract: Rapid advances in AI capabilities are causing benchmarks to saturate. Once several models approach perfect scores, a benchmark loses its power to differentiate between them, and with it its utility for informing AI safety cases and…
arXiv:2610.20675v1 Announce Type: new Abstract: Agents pair reasoning models with harnesses that organize model calls, execute tools, load skill instructions, and retain history in memory. Understanding agent performance requires linking execution conditions to outcomes,…
arXiv:2610.20845v1 Announce Type: new Abstract: Skill-policy co-evolution, in which an agent's policy model and its skill bank improve each other during training, is a promising paradigm for self-improving agents. Current evaluations score the co-evolved policy on the final bank…
arXiv:2610.20669v1 Announce Type: new Abstract: Reward-guided discrete diffusion enables test-time alignment with downstream objectives, but applying Sequential Monte Carlo (SMC) guidance at every denoising step is computationally expensive. Active guidance steps require…
arXiv:2610.20921v1 Announce Type: new Abstract: Most debate about releasing capable AI systems treats the choice as binary: keep the model behind an API, or publish the weights. That framing runs together two separate decisions. One is how capable a system may become. The other…
arXiv:2610.20780v1 Announce Type: new Abstract: In LLM-based multi-agent systems (MAS), a local mistake can propagate as other agents build on it, turning a correctable error into a failed task. Diagnosing completed runs or repeatedly re-executing them to search for corrections…
arXiv:2610.20809v1 Announce Type: new Abstract: A self-improving agent has to decide which new skills to keep, and in practice it keeps whatever earns a high reward, convinces a judge model, or compresses its library, even when the skill is wrong. That is fine until the answers…
arXiv:2610.20867v1 Announce Type: new Abstract: We present a general declarative framework for anticipatory reasoning about interactive motion in complex dynamic environments. Our focus is on (explainable) anticipatory space-time mental model formation at a semantic level to…
arXiv:2610.20782v1 Announce Type: new Abstract: The closed world asks AI for specialized competence and lets it manufacture the iid data its theory assumes: a fixed criterion licenses collecting examples until the task is covered. The open world asks for versatility --- taking…
arXiv:2610.20805v1 Announce Type: new Abstract: A reasoning trace retained at answer time plays two roles: it can raise accuracy, and it can compete with the current evidence when the two disagree. We disentangle these roles by crossing verified evidence versions with the traces…
arXiv:2610.20665v1 Announce Type: new Abstract: Large language model (LLM) agents can solve complex interactive tasks, yet reasoning and tool calls take time. In continuously changing environments, an otherwise correct action may arrive too late. We introduce RealtimeBench, with…
arXiv:2610.20836v1 Announce Type: new Abstract: Evaluator-guided reasoning presents a local control problem at every step: trust the current answer, return a fallback, or spend more inference. We formulate this choice as a compute-aware three-action controller, scoring terminal…
arXiv:2610.20951v1 Announce Type: new Abstract: Decision models answer typed questions about an input with one probability per option, computed in a single forward pass without generating text. Like the fast System 1 of dual-process theory, they answer at once, but their depth…
arXiv:2610.20685v1 Announce Type: new Abstract: Symbolic regression searches for an equation that explains observed data. Numerical error is commonly used to guide the search, but it alone can favor an accurate surrogate over a candidate with the right structure. We propose…
“ИИ станет хищной плесенью“. Почему сами разработчики считают его угрозой и как он уже вредит TengriNews
OpenAI's ChatGPT Atlas browser is gone, but now ChatGPT's Google Chrome extension is getting some upgrades that allow it to use the browser tabs you already have open. With the extension installed, OpenAI says ChatGPT can now view your browser tabs when you tag Chrome in…
✔️ OpenAI ведёт переговоры о покупке ИБ-компании Trail of Bits Обе компании уже работали вместе над проектом Patch the Planet, где модели OpenAI в связке с инструментами Trail of Bits автоматически искали и исправляли ошибки в открытом коде. Нью-йоркская Trail of Bits основана в…
🤖 Gemini 4 Argon научила гуманоида играть в Корнхол, просто посмотрев видео Модель проанализировала движение, восстановила физику броска, сама написала симуляцию и обучила гуманоида Apollo от Apptronik. В результате робот смог попасть в цель 7 раз подряд. Argon также набрала…
President Donald Trump posted a series of photos on Truth Social showing off renovations to the White House - but one of them appears to be modified with AI. The picture, which compares an old White House column to a "restored" one, has several discrepancies and was flagged by…
Powerful AI models continue to cause chaos, but their targets aren’t widely seeking justice. Instead, they’re being encouraged to adopt those same tools.
Anthropic предоставляет METR доступ к своим моделям для независимой оценки.
AWS is bringing multiple sessions to in San Jose. AWS Trainium is an AI chip designed from the ground up for next generation deep learning AI training and inference, and now PyTorch runs natively on AWS Trainium -- zero code changes. Same APIs, same debugging, same ecosystem.…