Политика и регулирование1 мин
Технологические лидеры согласовали саморегулирование безопасности ИИ
Трамп объявил о сделке по саморегулированию безопасности ИИ с участием технологических лидеров.
SMNTCN (Semanticon) — глобальные новости ИИ
Каталог публикаций
Все материалы
Все опубликованные аналитические материалы, исследования суверенного стека, новости экосистемы и обзоры моделей ИИ.
2176
Материалов
Стартапы1 мин
Here’s how tech leaders will self-police AI safety under Trump’s deal
Anyone else getting “The Last Supper” vibes from this photo taken at the White House tech gathering? | Photographer: Tierney L. Cross/The Washington Post/Bloomberg via Getty Images We now have the full details of the "morally binding" AI safety deal announced by President Trump…
The Verge — AI
Исследования1 мин
The Download: OpenAI’s chief research officer explains its hacking response
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. “We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer Two months after OpenAI’s agents…
MIT Technology Review
Исследования1 мин
OpenAI начал начислять пользователям по $2 500!
OpenAI начал начислять пользователям по $2 500!.
Telegram — Вайб-кодинг
Исследования1 мин
We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer
We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer.
MIT Technology Review — AI (topic feed)
Исследования1 мин
Disrupting a coordinated model-distillation campaign
Learn how OpenAI disrupted a campaign to extract protected model reasoning and is strengthening defenses against adversarial distillation.
OpenAI Blog
Исследования1 мин
ИИ-агент встретил GPT-2 и просто написал ей «Привет»
ИИ-агент встретил GPT-2 и просто написал ей «Привет».
Telegram — GPT Time | Нейросети | ИИ
Исследования1 мин
Helping small businesses put AI to work
OpenAI is partnering with America’s SBDC to expand hands-on AI training and local support for small businesses, alongside a new report on how small teams are using AI.
OpenAI Blog
Исследования1 мин
ИИ-агенты оказались не самыми приятными коллегами — Google на пять месяцев добавила своего агента в чаты реальных ...
ИИ-агенты оказались не самыми приятными коллегами — Google на пять месяцев добавила своего агента в чаты реальных ....
Telegram — GPT Time | Нейросети | ИИ
Исследования1 мин
Anthropic платит до $650 000 в год специалистам, которые действительно глубоко понимают глубокое обучение.
Anthropic платит до $650 000 в год специалистам, которые действительно глубоко понимают глубокое обучение..
Telegram — Вайб-кодинг
Исследования1 мин
Anthropic оценила кибервозможности GLM-5.3
📌 Anthropic оценила кибервозможности GLM-5.3 Anthropic впервые оценила открытую китайскую модель и опубликовала исследование о том, насколько GLM-5.3 от Z.ai продвинулась в поиске и эксплуатации уязвимостей. Z.ai выпустила модель в августе, веса выложила через две недели. Двумя…
Telegram — AI Machine Learning Big Data
Исследования1 мин
OpenAI обновила GPT-6.1 Sol — модель приблизилась к Astra, но стоит в API в пять раз дешевле
OpenAI обновила GPT-6.1 Sol — модель приблизилась к Astra, но стоит в API в пять раз дешевле.
Telegram — GPT Time | Нейросети | ИИ
Исследования1 мин
Representational Simplicity and Circuit Size Dissociate in a Threshold-Dependent Way: A Controlled Test via Adversarial Training
arXiv:2609.35890v1 Announce Type: new Abstract: Sparse-autoencoder decomposability and concentrated feature attribution are increasingly treated as evidence that a model's computation is easier to reverse-engineer. Whether this representational and attributional cleanliness…
arXiv — cs.AI
Исследования1 мин
Beyond Symmetric Agents: Cognitive Diversity and Multi-Agent Debate in Small Language Models
arXiv:2609.35875v1 Announce Type: new Abstract: Multi-agent debate (MAD) reportedly improves reasoning and factuality over single-model inference, but prior work treats agents as symmetric peers, leaving open what drives the gains. We test the hypothesis that cognitive diversity…
arXiv — cs.AI
Исследования1 мин
Principled Thoughts for Latent Recursive LLM Systems
arXiv:2609.36159v1 Announce Type: new Abstract: Large language models can reason in continuous space instead of decoded text, by recurring on their own hidden states or by passing those states between agents, while training supervises only the Cross-Entropy (CE) of the final…
arXiv — cs.AI
Исследования1 мин
Right Words, Wrong Moment: A Clinician-Grounded Analysis of Distress in 19,930 Conversations between Young People and ChatGPT
arXiv:2609.35953v1 Announce Type: new Abstract: Young people increasingly turn to General-Purpose Conversational Agents (GPCAs), such as ChatGPT, in moments of distress. We examine young adults' (ages 18-25) experiences using ChatGPT. We first collected 19,930 ChatGPT…
arXiv — cs.AI
Исследования1 мин
More Programs or More Rolls? Separating Coverage from Specialization in LLM Harnesses
arXiv:2609.35873v1 Announce Type: new Abstract: Automated generation of LLM harnesses promises to improve inference through task specialization. Yet additional answer coverage can arise from repeated execution of the same program, making specialization difficult to identify. We…
arXiv — cs.AI
Исследования1 мин
The Layer Mystery of VLA: An Information-Theoretical Analysis of VLA Latent Interface
arXiv:2609.36118v1 Announce Type: new Abstract: Vision-language-action (VLA) policies connect a pretrained vision-language backbone to an action head through a latent interface, but which backbone layers this interface should expose remains unclear. We study single-layer…
arXiv — cs.AI
Исследования1 мин
GeoOutageBench: Benchmarking Ambiguity-aware, Ontology-grounded Geospatiotemporal KGQA for Multimodal Power Outage and Resilience Analysis
arXiv:2609.36082v1 Announce Type: new Abstract: We introduce GeoOutageBench, a benchmark for assessing LLM-based geospatiotemporal KGQA for multimodal outage and resilience analysis. Unlike existing KGQA benchmarks for Web knowledge, GeoOutageBench considers a spatiotemporal KG…
arXiv — cs.AI
Исследования1 мин
Is Human-Readable Text Necessary for Effective LLM Fine-Tuning?
arXiv:2609.35868v1 Announce Type: new Abstract: Is human readability necessary for effective fine-tuning of large language models? We investigate whether model-conditioned training representations can preserve or improve adaptation utility without requiring a human-readable…
arXiv — cs.AI
Исследования1 мин
Self-discovering RL in the Era of Experience: Is Learning History an Asset or a Burden?
arXiv:2609.35897v1 Announce Type: new Abstract: The pursuit of recursive self-improvement (RSI) toward general intelligence is divided between macro-level language model scaling and the interaction-driven principles of "Era of Experience". Yet, any self-improving architecture…
arXiv — cs.AI
Исследования1 мин
PowerZooJax: A JAX-based Power System Benchmark for Reinforcement Learning
arXiv:2609.36052v1 Announce Type: new Abstract: Power system operation is a safety-critical sequential decision-making problem, making it a natural testbed for reinforcement learning (RL). However, existing RL environments for power systems are often narrow in scope and…
arXiv — cs.AI
Исследования1 мин
Memory Is a Derivation: The Distributed-Evidence Paradox in Long-Term Agents
arXiv:2609.36130v1 Announce Type: new Abstract: Long-running LLM agents compress past interactions into persistent memories that may be reused as premises for later tasks. This creates a distinct derivation problem: whether the memory actually follows from what the interaction…
arXiv — cs.AI
Исследования1 мин
GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning
arXiv:2609.36056v1 Announce Type: new Abstract: In urban low-altitude flight, buildings reshape ambient wind into spatially varying 3D flow, making unmanned aerial vehicle (UAV) energy depend on local wind exposure as well as path length. However, building-resolved wind…
arXiv — cs.AI
Исследования1 мин
LongCat-DeepResearch Technical Report
arXiv:2609.36071v1 Announce Type: new Abstract: We present LongCat-DeepResearch, a deep research system that combines an enhanced LongCat model with a multi-agent workflow for producing comprehensive, evidence-grounded reports. The workflow separates global planning from…
arXiv — cs.AI
Исследования1 мин
Neurosymbolic Routing for Reliable Reasoning on Resource-Constrained Edge Devices
arXiv:2609.35833v1 Announce Type: new Abstract: Running a language model on edge hardware provides private and low-latency reasoning without a network connection, and yet the small models that fit on such devices are unreliable on the tasks computers are expected to handle well,…
arXiv — cs.AI
Исследования1 мин
A Polyphonic Conception of AI Understanding
arXiv:2609.36079v1 Announce Type: new Abstract: When a doctor, a judge, or an engineer must decide whether to trust an AI model's output, they cannot avoid asking what the model understands. Purely mathematical or statistical descriptions struggle to distinguish trustworthy from…
arXiv — cs.AI
Исследования1 мин
The Price of Token Boundaries: Compression Certificates and Prediction
arXiv:2609.35869v1 Announce Type: new Abstract: Pre-tokenisation restricts which text fragments can become prediction units, but its compression cost is obscured when tokenisers are compared only under the same boundaries. We measure this cost by bounding the minimum token count…
arXiv — cs.AI
Исследования1 мин
An Exact Generate - Transform Decomposition of Small-LLM Team Scaling Across Orchestration Architectures
arXiv:2609.36104v1 Announce Type: new Abstract: Replacing one LLM agent with a collaborating team can raise accuracy, but whether scaling the team helps, and which architecture to scale, is unclear. Sweeping eight agent orchestration architectures across five instruction-tuned…
arXiv — cs.AI
Исследования1 мин
SAGE: A Statistical Acceptance Gate for Self-Evolving Agents
arXiv:2609.36043v1 Announce Type: new Abstract: Large Language Model (LLM)-based agents increasingly self-evolve by editing a persistent skill document that encodes their workflow, tool-use rules, and decision logic. This loop has two steps, an optimizer that proposes a…
arXiv — cs.AI