Full-text search across 70383 articles. Combine with topic and date filters; results sorted by relevance.
IBM and Red Hat have found and fixed more than 400 previously unknown vulnerabilities in widely used Java libraries through Lightwell, their program for patching open source code that companies already run in production.…
Researchers introduced U-Space, a low-dimensional subspace that makes a language model's evolving uncertainty measurable and interpretable without correctness labels, repeated generations, or training, according to the a…
Researchers established a comprehensive benchmark for X-ray radiology report generation on the CheXpert Plus dataset, which was released without baseline implementations or evaluation results, and proposed a new framewor…
A new arXiv paper (2610.09152v1) introduces sk-bench, a native-first Slovak benchmark comprising 30 datasets and 33 scored task variants across ten skill categories, and reports that the best open-weights model trails pr…
A new arXiv paper (2610.08835v1) proposes a Gated Cross Attention (GCA) framework that uses LLM-generated explanations from original news articles as stable background knowledge to detect fake news that has been emotiona…
A paper on arXiv (2610.09145v1) reports that noising conditioning prompt tokens during training, rather than keeping them clean, improves continuous diffusion language models on combinatorial reasoning tasks. The single-…
A developer built DeskVNCViewer, a native VNC/RDP client that lets an AI agent drive a remote desktop session while keeping passwords and credentials out of the agent's reach. The agent can view, click, type and navigate…
Helpin, an open-source shared support and project workspace with configurable AI agents and MCP integrations, shipped its Community 0.2 beta release, with community application code licensed AGPL-3.0-only and public SDKs…
Anthropic released Claude Haiku 5.5, which the company calls its cheapest, fastest, and most capable small model to date, priced at $0.10-$0.50 per million input tokens and $0.50-$2.50 per million output tokens. Anthropi…
Mistral AI has deprecated Pixtral 12B, the text-and-image model released on 2024-09-01 with a 128k context window and 128k max output, according to a provider announcement tracked by models.dev. No retirement date has be…
Alphabet's Google and Unity Software announced Playground on Oct 7, a platform that lets people 18 and older generate and customize video games from natural language text prompts with no coding required. A new expanded c…
A developer built Janela, an open-source local weather-window planner that pairs a Spring Boot API and Open-Meteo forecast data with a deterministic Java WindowScorer to rank the best hours for outdoor activities, then u…
Anthropic resolved an issue on October 7 at 21:23 UTC that had incorrectly paused some organizations for reaching their spend limit, causing refused requests across the Claude API, Claude.ai, Claude Code, Claude Security…
A developer published logo-design-skill, an MIT-licensed repository that packages a staged logo-design workflow as an Agent Skill for Claude Code, Gemini CLI, Codex CLI and other compatible agents. The skill guides an ag…
A developer built Touch Grass AI, an offline-first edge expedition planner that runs Google's Gemma 2 (2B) instruction-tuned model locally via Hugging Face Transformers to generate nature outing plans without cloud APIs.…
Anthropic released Claude Haiku 5.5, upgrading its smallest and cheapest Claude model from version 4.5 and marking the third Claude 5.5 upgrade in a month after Opus 5.5 and Sonnet 5.5. Anthropic said Haiku 5.5 costs aro…
A benchmark of vLLM 0.30.0, SGLang 0.5.20, and TensorRT-LLM 1.2.1 on a single H100 SXM GPU with 8 CPU cores found no host-bound regime across Qwen 3 0.6B-32B models at batch sizes from 1 to 256, with CUDA graphs enabled,…
OpenAI's GPT-6.1 Sol supports async tool calling, letting an application hand off a slow task or human question while the model continues independent work, but the API only removes the model-side wait and does not run jo…
Docker released docker-agent, a docker CLI plugin that lets developers build, run, and share AI agents through declarative YAML configuration with no code required. The tool supports multi-agent orchestration, any MCP se…
A developer built Kriterio, a Chrome extension that checks web listings against user-defined rules and flags them green, amber or red with a stated reason. After finding that an LLM misread prices and invented return pol…