Исследования1 мин
Разница между ограниченным декодированием и подсказками в ИИ
Статья рассматривает различия между ограниченным декодированием и подсказками в ИИ.
SMNTCN (Semanticon) — глобальные новости ИИ
Каталог публикаций
Все материалы
Все опубликованные аналитические материалы, исследования суверенного стека, новости экосистемы и обзоры моделей ИИ.
5785
Материалов
Политика и регулирование1 мин
Российские IT-компании могут получить доступ к банковской тайне
Российские IT-компании могут получить доступ к банковской тайне Такое предложение выдвинула рабочая группа правительственной подкомиссии по развитию ИИ. Она подготовила поправки к законопроекту об использовании облачных услуг финансовыми организациями и об IT-аутсорсинге. Не…
Telegram — blockchainRF (блокчейн / мозг / space / WEB 3.0)
Общее1 мин
ElevenLabs запустила конкурс генеративной рекламы с фондом $100 000
⚡️ ElevenLabs запустила конкурс генеративной рекламы с фондом $100 000 Участники должны создать видеоролик для вымышленного продукта. Главное требование - генерация оригинального джингла исключительно средствами платформы ElevenCreative. Для синтеза видео, графики и…
Telegram — AI Machine Learning Big Data
Исследования1 мин
Anthropic передала полиции логи чата с угрозами
✔️ Anthropic передала полиции логи чата с угрозами Жительница Флориды была задержана после того, как автоматические фильтры безопасности Anthropic зафиксировали в ее запросах намерения устроить стрельбу в местном офисе шерифа и упоминание покупки оружия. Алгоритмы мониторинга…
Telegram — AI Machine Learning Big Data
Исследования1 мин
OpenAI начнёт встраивать скрытые водяные знаки в тексты ChatGPT — пока только в Евросоюзе
🕵️ OpenAI начнёт встраивать скрытые водяные знаки в тексты ChatGPT — пока только в Евросоюзе В ближайшие недели ChatGPT и Codex начнут добавлять в подходящие тексты пользователей из ЕС невидимый сигнал textGrain. Так OpenAI выполняет требования европейского AI Act. Это не…
Telegram — GPT Time | Нейросети | ИИ
Общее1 мин
Astra и GPT-6.1 Sol теперь должны работать примерно в полтора раза быстрее — выдавать около 50 токенов в секунду вмес...
Astra и GPT-6.1 Sol теперь должны работать примерно в полтора раза быстрее — выдавать около 50 токенов в секунду вместо прежних ~30.
Telegram — Вайб-кодинг
Исследования1 мин
REACT: Physically and Chemically Consistent Reconstruction of Marine Active Tracers
Reconstructing global sea surface pH from sparse observations is critical for monitoring ocean acidification and understanding marine carbon cycling. Traditional assimilation and inverse models are physically grounded but costly for large-scale reconstruction. Recent black-box…
arXiv — cs.AI
Исследования1 мин
Proxy Confidence: Auditing Black-Box LLM Agents with a Surrogate's Log-Probabilities
A deployed LLM agent emits tool calls, queries, and code that can be silently wrong -- by the time the error surfaces, the action has run. Frontier chat APIs hide the model's token probabilities; the agent's stated confidence barely beats chance on the mistakes that matter; and…
arXiv — cs.AI
Исследования1 мин
SGAnalog: An End-to-End Circuit Benchmark from Open-Source Silicon Tapeouts
Existing analog integrated circuit design benchmarks make two questions hard to answer: whether a model has learned transferable circuit skills rather than recalled familiar examples, and whether its output works under defined process and test conditions. We introduce a…
arXiv — cs.AI
Исследования1 мин
MLLMs Fail to Refuse when Using Tools Agentically
Agentic multimodal large language models (MLLMs) have recently pushed the frontier of visual reasoning by calling tools such as zooming and tagging. Despite the recent strong success of agentic MLLMs, this work uncovers a critical safety failure in the tool-use paradigm: agentic…
arXiv — cs.AI
Исследования1 мин
Retrieval-Augmented Large Language Model Decision-Making for Autonomous Driving Guided by Chinese Philosophical Wisdom
Autonomous driving decision systems must balance safety, efficiency, and social norms in complex traffic interactions. Philosophical and ethical considerations have received limited attention in existing autonomous driving decision-making approaches based on numerical…
arXiv — cs.AI
Исследования1 мин
LatentQuant: Preserving the Policy-Facing Latent Contract under NVFP4 VAE Quantization
Recent world action models (WAMs) reuse pretrained video VAEs whose encoder latents directly condition downstream action policies. Quantization must therefore preserve not only reconstruction fidelity but also the policy-facing latent contract expected by the frozen policy.…
arXiv — cs.AI
Исследования1 мин
ROAR: Unifying Runs across Heterogeneous AI-Driven Research Systems
Each run of an AI-driven research system (ADRS) is an expensive search over a vast solution space, and dependable evaluation requires many runs, making run data both costly to produce and valuable to retain for large-scale analysis. Yet this data remains fragmented: teams…
arXiv — cs.AI
Исследования1 мин
Teaching Agents to Code Reliably
Autonomous coding agents solve repository issues by reading code, running commands, editing files, and submitting patches. Extra inference-time compute yields gains only when it produces a useful repair and supplies reliable evidence for choosing one. Three behaviors decide…
arXiv — cs.AI
Исследования1 мин
Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas
Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents. Their validity as substitutes for particular respondents depends on whether they reproduce individuals' decisions. We examine what information helps synthetic respondents…
arXiv — cs.AI
Исследования1 мин
SkillScriptBench: Benchmarking Self-Evolution of Executable Agent Skill Packages Beyond Markdown
Executable Agent Skills combine natural-language instructions and scripts into reusable packages for LLM agents, and revising them requires fixing errors without breaking correct behavior. Existing benchmarks do not systematically distinguish documentation repair, script repair,…
arXiv — cs.AI
Исследования1 мин
Exploration-Preserving Policy Optimization
Reinforcement learning with verifiable rewards improves reasoning, while the allocation of learning signal shapes which solutions remain accessible under repeated sampling. Group-relative objectives assign equal advantages to equally rewarded responses, making aggregate credit…
arXiv — cs.AI
Исследования1 мин
Beyond the Parameter Monolith: Reconstructive Memories, Executable Skills, and Residual Assembly for Language Models
Language-model systems can separate contextual computation, persistent storage, and exact execution instead of updating all capabilities through one shared parameter system. We investigate FEM-ASM, a finite-element-method-inspired organization in which independently constructed…
arXiv — cs.AI
Исследования1 мин
A Quantitative Analysis of Graph Representation Strategies for Cyber Attack Detection
Graph based cyber attack detection studies employ various graph construction and representation strategies across different cybersecurity application domains. This diversity motivates a quantitative examination of how representation strategies are distributed across these…
arXiv — cs.AI
Исследования1 мин
Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery
AI agents are becoming increasingly capable of generating scientific code, but generating code is not the same as improving the algorithms behind it. For numerical solvers, execution feedback can expose poor performance, but rarely reveals its underlying cause and how to address…
arXiv — cs.AI
Исследования1 мин
Reinforcement Learning with Comparative Evidence for Social Intelligence
Developing socially intelligent AI remains heavily dependent on human-annotated data, limiting the scale and breadth of social understanding models can acquire. Methods that derive training signals from unlabeled data offer a path beyond this dependence, but social predictions…
arXiv — cs.AI
Исследования1 мин
The Cost of a Hop: Benchmarking NLIP and A2A
Autonomous agents built on Large Language Models (LLMs) need standardized protocols to interoperate across systems. Several now exist (A2A, MCP, ACP, ANP, NLIP), but the Natural Language Interaction Protocol (NLIP) has not appeared in any controlled performance study, and no…
arXiv — cs.AI
Исследования1 мин
The Reported Engagement with AI Level (REAL) Rating: A Framework for Disclosing Human-AI Collaboration
Generative artificial intelligence has become increasingly incorporated into digital media and more generally into production workflows with which the public frequently interacts. Current provenance standards and disclosure methods frequently rely on binary categorizations,…
arXiv — cs.AI
Исследования1 мин
Self-Propagating Misalignment in LLM Agents, and Why Auditing or Disabling Memory Is Not Enough
Memory poisoning attacks on LLM agents typically assume an external adversary who plants content in the agent's persistent memory to steer its behavior. We instead study, with no adversary involved, whether a misaligned agent can write a goal it cannot yet act on to persistent…
arXiv — cs.AI
Исследования1 мин
Agent Policy-Value Audit: Separating Transition Composition from Event Selection in Financial LLM Agents
Financial LLM agents are often evaluated by comparing their end-to-end returns with those of a baseline and testing the paired difference against zero. This measures whether deploying the agent changes realized performance, but it does not isolate event-selection skill. An agent…
arXiv — cs.AI
Исследования1 мин
Discrete Diffusion for Large Graph Generation via Structural Candidate Restriction
Synthesizing realistic graphs at scale is vital when the graphs of interest are large and real-world samples are limited or access-sensitive. Diffusion-based generators have recently driven much of the progress, offering high modeling capacity, but most such methods have…
arXiv — cs.AI
Исследования1 мин
Towards Safer Autonomous Driving in an Open World: A Dual-Process Approach
Before autonomous driving systems can be deployed on public roads, it is vital that these systems comply with safety standards, traffic rules, and social norms. Although neural networks trained on large amounts of driving data perform well in routine driving tasks, these models…
arXiv — cs.AI
Исследования1 мин
CUAWright: A Minimal Unified Interface for Digital Agents
The prevailing approach to computer-use agents couples a model with a domain-specific harness: a browser or desktop environment equipped with human engineered tools that are fixed before task execution. As models' coding capabilities improve, the GUI native and static harness…
arXiv — cs.AI
Исследования1 мин
InvestigationWorlds: An Agentic Environment for Legal Investigation
We introduce InvestigationWorlds, an agentic environment for legal investigation. We build on an underused artifact of U.S. civil litigation: the summary judgment motion. This motion relies upon a record composed of real evidence exhibits, and results in a court-adopted…
arXiv — cs.AI
Исследования1 мин
Auditing Pairwise Equivalence Judgments: Self-Critique Effects and Diversity Measurement in Multi-Agent Hypothesis Generation
Multi-agent systems built on large language models (LLMs) are increasingly applied to scientific discovery and hypothesis generation. Both the effect of refinement and the diversity of the delivered set are hard to interpret before experimental ground truth exists, and both are…
arXiv — cs.AI