在 ChatGPT 中，您可以使用 PIN 码或 Face ID 隐藏对话
🔒 在 ChatGPT 中，您可以通过 PIN 码或 Face ID 隐藏对话 OpenAI 已开始逐步推出“锁定聊天”功能，该功能可让您保护 ChatGPT 中的个人对话免受陌生人的侵害。用户可以隐藏常规历史记录中的个人或机密聊天记录，并且...
Telegram — BlueScreen.kz
主题
Research
2224материалов
RUЕвропейские пионеры используют модели NVIDIA Earth-2 для прогноза погоды
Модели NVIDIA Earth-2 помогают прогнозировать погоду и климат в Европе.
SMNTCN (Semanticon) — глобальные новости ИИ
RUNVIDIA трансформирует европейскую розницу с помощью ИИ и роботов
NVIDIA внедряет открытые модели и физический интеллект в европейскую розницу.
SMNTCN (Semanticon) — глобальные новости ИИ
哈萨克斯坦的小学生通过检查人工智能错误的项目在巴厘岛举行的一项著名比赛中成为最好的学生
来自阿拉木图的两名学童在巴厘岛举行的 IRCYS 2026 比赛中获得银牌。他们的 PlanLens 系统发现人工智能代理计划中的错误。
DigitalBusiness.kz
克劳德现在建议该怎么做：Anthropic 推出了主动式人工智能代理
📎 Claude 现在亲自建议该怎么做：Anthropic 推出了主动式 AI 代理 该公司推出了 Claude Duet 的测试版本，这是一种基于 Claude Opus 5.5 模型的主动式 AI 代理。与常规聊天机器人不同，它不会等待持续的命令，而是独立地......
Telegram — BlueScreen.kz
Research🇰🇿 哈萨克斯坦专题：Anthropic представляет Claude Duet — нового AI-агента-конкурента Muse#OpenAI#Anthropic1 分钟
ENAnthropic introduced Claude Duet, an autonomous AI agent
Anthropic has introduced a new AI-agent called Claude Duet. It is designed to make everyday work easier for professionals and small business owners: they can schedule meetings, make financial forecasts and develop web applications, reports Tech-news. Claude Duet's main…
Tech-news.kz
Research专题：Anthropic представляет Claude Duet — нового AI-агента-конкурента Muse#OpenAI#Anthropic1 分钟
Anthropic 推出了 Claude Duet，一款无需永久命令即可工作的个人 AI 代理
⚡️ Anthropic 推出了 Claude Duet，这是一款无需永久命令即可工作的个人人工智能代理。新服务建立在Claude Opus 5.5的基础上，将成为OpenAI和Grok Bot的Dots的竞争对手。 Duet 在云中运行，使用浏览器和连接的应用程序，并执行...
Telegram — GPT Time | Нейросети | ИИ
Kapital.kz
Anthropic 决定发布基于极其强大的 Opus 5.5 的 Claude Duet，来挑战 OpenAI Dots 和 Grok Bot。
Anthropic 决定发布基于极其强大的 Opus 5.5 的 Claude Duet，来挑战 OpenAI Dots 和 Grok Bot。 • 适用于Claude Pro 计划（Dots 需要每月支付100 美元订阅ChatGPT Pro）。 • 运行Opus 5.5，被称为世界上最好的型号。 • 有…
Telegram — Вайб-кодинг
ENForging More Tasks into Better Harnesses: Agentic Harness Evolution via Successive Scale Racing
arXiv:2610.28815v1 Announce Type: new Abstract: Agentic harness evolution optimizes an agent's instructions, tools, and control flow by using its task execution experiences as evolving evidence. While more evolution tasks should intuitively improve performance, our preliminary…
arXiv — cs.AI
ENSpecEvo: Speculative Evolution with Large Language Models for Cost-Efficient Scientific Discovery
arXiv:2610.28706v1 Announce Type: new Abstract: LLM-driven evolutionary search is a promising paradigm for scientific discovery, but existing methods remain costly and slow, often applying expensive reasoning uniformly while underusing evidence from failed, redundant, and…
arXiv — cs.AI
ENScaling in Games: Continued Pre-Training For Embodied Agents In Diverse Virtual Worlds
arXiv:2610.28698v1 Announce Type: new Abstract: Scaling data, diversity and model capacity has improved embodied learning in physical environments, where diverse experiences remain grounded in shared physical laws. Whether these gains extend to virtual worlds with different…
arXiv — cs.AI
ENA Small Model Asks Better: Fine-Tuning Query Clarification for Retrieval-Augmented Generation
arXiv:2610.28686v1 Announce Type: new Abstract: Query clarification is essential for retrieval-augmented generation (RAG) systems that face ambiguous user queries. We consider multi-turn grounded conversational scenarios where a user query is either ill-formed or ambiguous in…
arXiv — cs.AI
Research
KZ·AI FA39Automating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild?
ENAutomating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild?
arXiv:2610.28763v1 Announce Type: new Abstract: Frontier coding agents are increasingly framed as end-to-end automated researchers, and ML for science offers a demanding test of that promise: improving a scientific model often requires domain insight beyond routine ML…
arXiv — cs.AI
ENDecision language models as affordable and rapid alternatives to LLM judges of clinical generative AI
arXiv:2610.28826v1 Announce Type: new Abstract: Automated judges of clinical artificial intelligence (AI) outputs are necessary, but large language model (LLM) judges are costly and slow at scale. We compared Jev, a decision language model that returns structured judgments, with…
arXiv — cs.AI
ENHidden Messages Between Diffusion Language Model Agents: Causal Evidence for Latent Communication
arXiv:2610.28905v1 Announce Type: new Abstract: Diffusion language models form answers through a sequence of revisable continuous states, but multi-agent systems usually discard those states and make agents communicate only after decoding text. We ask whether one independent…
arXiv — cs.AI
ENDiffResearch: Accelerating Deep Research with Diffusion Language Models
arXiv:2610.28909v1 Announce Type: new Abstract: Deep research systems iteratively plan, gather evidence, synthesize, and revise, making end-to-end latency a key bottleneck for interactive use. Almost all production systems are built on autoregressive (AR) language models and…
arXiv — cs.AI
ENSimply Logical -- The First Three Decades
arXiv:2610.28912v1 Announce Type: new Abstract: This paper charts the evolution of the Prolog textbook Simply Logical -- Intelligent Reasoning by Example from print with runnable programmes on a 3.5-inch diskette published in 1994, via various intermediate online versions, to…
arXiv — cs.AI
ENBack Up Before You Regret: Measuring Backup Awareness and Automating Backup in Coding Agents
arXiv:2610.28934v1 Announce Type: new Abstract: Coding agents now complete tasks autonomously, modifying real-world state such as artifacts, software environments, and configurations along the way. Yet they show limited backup awareness: recognizing that an action will modify…
arXiv — cs.AI
ENSelf-Improving Agents with World Model Predictive Paths
arXiv:2610.28945v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly expected to interact with environments and improve their decisions during deployment. However, how anticipating future outcomes and drawing on accumulated interaction experience…
arXiv — cs.AI
ENGRAIL: From Saliency to Semantics in Time-Series Explanation
arXiv:2610.29036v1 Announce Type: new Abstract: Explaining a time-series model requires both locating the evidence and extracting meaningful knowledge from it. Yet a single timestep is not a semantic unit: the meaningful patterns live across many samples, not in any one of them.…
arXiv — cs.AI
ENSpeedrunning Relational Foundation Model Pretraining
arXiv:2610.28855v1 Announce Type: new Abstract: Progress on foundation model pretraining is gated by who can afford a trial, and relational pretraining is no exception. Recipes are developed inside individual labs, reported under private protocols, and do not accumulate.…
arXiv — cs.AI
ENTo Manipulate or Not: Opinion Dynamics in LLM Interactions
arXiv:2610.28875v1 Announce Type: new Abstract: Large language models increasingly take part in extended conversations where they can both shape and be shaped by another party's views, yet most persuasion research measures only one side of that exchange. We introduce a…
arXiv — cs.AI
ENSketch2Sim: an end-to-end modular framework for hand-drawn circuit analysis
arXiv:2610.28836v1 Announce Type: new Abstract: This study proposes a sketch-to-simulation system to solve handwritten circuit diagrams. The system converts raw images into a Simulation Program with Integrated Circuit Emphasis (SPICE)-compatible netlist and automatically…
arXiv — cs.AI
ENFig: Automated Multiagent Behavior Elicitor
arXiv:2610.28891v1 Announce Type: new Abstract: Many social evaluations for multiagent systems study behaviors as they emerge from environments instead of systematically intervening in the environment to elicit such behaviors. We introduce Fig (Finding Interactions in Games), an…
arXiv — cs.AI
END-Skill: Distilling Teacher Knowledge through Agent Skills
arXiv:2610.28914v1 Announce Type: new Abstract: Frontier language-model agents excel at complex tasks, but high inference costs make transferring their capabilities to smaller models essential. Traditional knowledge distillation achieves this through parameter updates, requiring…
arXiv — cs.AI
ENBehavior Beyond Text: Evaluating Refusal Transfer Failure in LLM Agents
arXiv:2610.28982v1 Announce Type: new Abstract: While large language models are increasingly deployed as tool-calling agents, their safety evaluations still focus primarily on their ability to refuse harmful requests in text. We investigate whether the refusal behavior in text…
arXiv — cs.AI
ENPerMIT: Securing Persistent Memory for LLM Agents via Machine-Checked Integrity Theorems and Scalable Community Trust
arXiv:2610.28987v1 Announce Type: new Abstract: Persistent memory is a largely undefended surface for injection and poisoning, and the same risk limits memory sharing among agents. We present PerMIT, a memory architecture with provable integrity guarantees under an explicit…
arXiv — cs.AI
ENAgentPolicyForge: From Executable Environments to Policy Benchmarks and Adaptive Guards
arXiv:2610.28989v1 Announce Type: new Abstract: Customized agent deployments define policies over tools, resources, and execution conditions. Providing these policies in the task agent's prompt does not reliably prevent violations. Training a guard requires examples of tool…
arXiv — cs.AI
ENAn Evidence-Driven Multi-Agent Compound AI System for Head and Neck Proton Radiotherapy Planning
arXiv:2610.29010v1 Announce Type: new Abstract: Radiation treatment planning requires balancing competing objectives to satisfy clinical criteria while accommodating physicians' preferences. This study aims to develop an evidence-driven, multi-agent compound AI system that…
arXiv — cs.AI
ENGenXR-Bench: Benchmarking Coding Agents for Interactive XR Development
arXiv:2610.29014v1 Announce Type: new Abstract: Coding agents are rapidly advancing from single-file generation to long-horizon repository-scale engineering, yet the evaluation of their capabilities to author interactive extended reality (XR) applications remains an open…
arXiv — cs.AI
ENInheritance as an Admission Test for LLM Prompt Merge Operators
arXiv:2610.29023v1 Announce Type: new Abstract: Large language model (LLM) agents are often improved by iterating until a fixed verifier is done. Once that verifier saturates, the loop has nothing left to climb. We instead score agents against the current population, and a…
arXiv — cs.AI
ENAutoFraming: Synthesizing Dynamic Frames of C Code with LLMs
arXiv:2610.29046v1 Announce Type: new Abstract: Large language models (LLMs) have shown considerable promise in synthesizing formal specifications from source code. However, existing approaches primarily focus on functional contracts and loop invariants, while largely neglecting…
arXiv — cs.AI
ENDoes Every Instance Count Equally? Order Effects in LLM Multi-Instance Processing
arXiv:2610.29057v1 Announce Type: new Abstract: Users increasingly rely on large language models (LLMs) to analyse collections of independent inputs, such as product reviews or social media posts, and return a single aggregated answer. This setting requires multi-instance…
arXiv — cs.AI
CLM-8B：用于选择代理操作的开放且快速的模型
🌟 CLM-8B：用于选择代理动作的开放且快速的模型 斯坦福大学与 NVIDIA 合作，公开了对比语言模型课程的第一个模型。这是 Jev 的开放模拟，由前研究员的 TypeSafe AI 初创公司于 9 月 15 日提出......
Telegram — AI Machine Learning Big Data
Anthropic 发布了 Claude Duet：第一个给你写信的私人经纪人
🤖 Anthropic 发布了 Claude Duet：第一个给你写信的私人经纪人。克劳德已经有了特工，但他们必须启动并等待。 Duet 的工作方式有所不同：您将一个任务放入类似信使的线程中，它在云计算机的后台执行，并在...时写入。
Telegram — AI Machine Learning Big Data
梦想实验室提出了一个控制陌生机器人的模型
✔️ Dream Labs 提出了一个控制陌生机器人的模型 来自 GR00T 机器人项目团队的四位前 NVIDIA 研究人员通过他们的初创公司走出了隐身状态，并开放了 Dream-1 模型的公共 API。该公司由 Joel Jang 领导，他领导了 GR00T……
Telegram — AI Machine Learning Big Data
RUAnthropic RELEASED RSI. Singularity Initiated! The Start of Recursive Self-Improvement! AI IS CRE...
Всем привет. В этом выпуске самые главные свежие новости об искусственном интеллекте и технологиях. Сооснователь Anthropic Джек Кларк заявил, что технологическая сингулярность уже наступила, а искусственный интеллект начал превосходить учёных и исправлять их ошибки.…
YouTube — AI Flash News
Google Cloud 将开放对 OpenAI AI 模型的访问
OpenAI 模型即将集成到 Gemini AI 代理中
Kapital.kz
RUФреймворк адаптивного агентного интеллекта использует Strands Agents
Фреймворк адаптивного агентного интеллекта использует Strands Agents для оптимизации AI-агентов.
SMNTCN (Semanticon) — глобальные новости ИИ
RUУченые улучшили ИИ-агентов, превращая научные статьи в код
Исследователи создали систему ScholarEvolve для оптимизации ИИ-агентов без переобучения, улучшив их производительность.
SMNTCN (Semanticon) — глобальные новости ИИ
RUAnthropic запускает Claude Duet, нового ИИ-агента для бизнеса
Anthropic представила Claude Duet, нового ИИ-агента для бизнеса, доступного в приложении.
SMNTCN (Semanticon) — глобальные новости ИИ
RUОбучение Qwen3 на AWS Trainium с использованием torchtitan
AWS Trainium позволяет обучать Qwen3 с использованием torchtitan и Hugging Face.
SMNTCN (Semanticon) — глобальные новости ИИ
Research专题：Anthropic представляет Claude Duet — нового AI-агента-конкурента Muse#OpenAI#Anthropic1 分钟
Anthropic 推出 Claude Duet 对标 Muse，推出新 AI 智能体
Anthropic 刚刚宣布推出 AI 智能体 Claude Duet，以竞争 Meta 的 Muse 和 OpenAI 的 Dots。据 Anthropic 称，这款面向专业人士和小企业主的新智能体将以「研究预览版」形式发布，并作为独立应用在 iOS 上提供……
The Verge — AI
OpenAI 的 GPT-5.6 Sol 与 Google 的 Gemini 4 Argon 发现影响数百万网站的关键 PHP 漏洞
OpenAI 的 GPT-5.6 Sol 作为“Patch the Planet”计划的一部分，发现了一个可能影响数百万网站的关键 PHP 漏洞。安全公司 Wiz 独立利用 Google 的 Gemini 4 Argon 发现了同一漏洞。OpenAI 警告，同样的 AI 能力……
The Decoder
Research1 分钟
RUИспользование ИИ для улучшения материнского ухода с помощью УЗИ
Исследование Google Research демонстрирует, как ИИ может улучшить материнский уход.
SMNTCN (Semanticon) — глобальные новости ИИ
NVIDIA Kumo Relational 在关系数据上超越有监督机器学习，无需训练
NVIDIA 在 Hugging Face 上的博客文章
Hugging Face Blog
RUClaude разработал белки, которые могут помочь в создании противоядия от укусов скорпионов
🦂 Claude разработал белки, которые могут помочь в создании противоядия от укусов скорпионов Исследователи использовали Claude для проектирования белков, способных связываться с опасными токсинами скорпионов и потенциально блокировать их действие. Всего за 64 часа ИИ создал…
Telegram — GPT Time | Нейросети | ИИ
PHP 末日：OpenAI 和 Google 的 Wiz 发现古老漏洞，引发互联网恐慌
这是一种被 AI 发现并预防的安全问题，但也因 AI 而成为高风险。
TechCrunch — AI
Google Cloud 现提供 OpenAI 模型，与 Gemini 和 Claude 并列，成为唯一同时提供这三大模型的主要云服务商。
Google Cloud 成为首家在单一平台上提供 Gemini、Claude 和 OpenAI GPT 模型的主要云服务商。发布时，GPT-6 Astra、GPT-6.1 Sol 和 GPT-6 Luna 以私有预览版面向企业客户提供。文章《Google Cloud 现提供 OpenAI 模型……》
The Decoder
Hugging Face Blog
RUDeel запускает модели Pangaea AI для HR и зарплат, снижая затраты
Deel представила модели Pangaea AI, которые обещают снизить затраты на HR и зарплаты.
SMNTCN (Semanticon) — глобальные новости ИИ
RUСтартап Recursive разрабатывает модель с рекурсивным самоулучшением
Стартап Recursive представил модель Flint, демонстрирующую рекурсивное самоулучшение ИИ.
SMNTCN (Semanticon) — глобальные новости ИИ
DeepMind创始人哈萨比斯将AGI时间表从五年缩短至两到三年
德米斯·哈萨比斯不断提前其AGI时间表。这位诺贝尔奖得主兼前Google DeepMind首席执行官现在预计人工通用智能将在两到三年内实现，较今年早些时候预测的三到五年有所提前。文章《DeepMind创始人哈萨比斯缩短AGI时间表……》
The Decoder
OpenAI 与谷歌正式宣布合作。
OpenAI 与谷歌正式宣布合作。 OpenAI 模型很快将在 Google Antigravity 和 Gemini Agent 中可用。 Google 已经提供了对 Gemini 和 Claude 模型的访问，现在将添加 GPT 模型。因此，反重力用户...
Telegram — Вайб-кодинг
TorchNeuron 性能分析（第 1 部分）：在 AWS Trainium 上的系统级性能分析。
Amazon 在 Hugging Face 上的博客文章
Hugging Face Blog
Wired — Artificial Intelligence
Research专题：Sam Altman: RSI, AI Consciousness, OpenAI's New Hardware, Compute, and more!#OpenAI#Сэм Альтман1 分钟
ENSam Altman: RSI, AI Consciousness, OpenAI's New Hardware, Compute, and more!
Do you believe today's models are conscious? >> I want to be clear. No one really knows. Can I say what I don't like about the whole framing? >> Should it ever be allowed to not answer or disengage with me? I think this is a technology that wants a new form factor. Some lab says…
YouTube — Matthew Berman
ENJensen Huang’s Vision for Europe Runs on Plug-In Data Centers
At Nvidia’s GTC in Berlin, Huang unveiled the company’s new plug-in data center—the DSX Pod—as the key to unlocking Europe’s AI infrastructure lag. Oh, and autonomous humanoid robotics.
Wired — Artificial Intelligence
Startups
KZ·AI 1F9AБывший глава Google владеет примерно 1% Anthropic — его доля может стоить $20 млрд
RUБывший глава Google владеет примерно 1% Anthropic — его доля может стоить $20 млрд
💰 Бывший глава Google владеет примерно 1% Anthropic — его доля может стоить $20 млрд По данным Bloomberg, Эрик Шмидт вложил десятки миллионов долларов в Anthropic ещё в 2021 году, когда стартап оценивался всего в $461 млн. Теперь ему, предположительно, принадлежит около 1%…
Telegram — GPT Time | Нейросети | ИИ