Full-text search across 70843 articles. Combine with topic and date filters; results sorted by relevance.
A developer built a local lab demonstrating the operational differences between running an MCP server on a laptop and hosting it behind a gateway for multiple clients and tenants. The project implements bearer-token auth…
A new arXiv paper (2609.10787v1) introduces "Shedding Light," a benchmark for evaluating whether generative image models understand lighting in a physically accurate manner. The benchmark uses a multi-illumination datase…
A prompt-based approach to minimal-edit grammatical error correction reached an F0.5 score of 78.32 on the BEA-2019 test set, closing the gap to the fine-tuned single-model state of the art (Staruch et al., 2025) to 0.38…
A new arXiv paper (2609.10830v1) reports that language models' apparent memory of training data is largely confounded, using exact duplication counts from the published pretraining corpora of OLMo-2 and Pythia. Across fi…
Researchers introduced LLM-Anchored Paralinguistic Enrichment (LAPE), a method that combines LLM-derived linguistic representations with paralinguistic speech cues to detect Alzheimer's disease, achieving state-of-the-ar…
Researchers introduced CityPlanner, a sandbox-agent framework for executable urban planning, detailed in arXiv paper 2609.09578v1. CityPlanner uses UrbanSandbox, a unified file-based environment where agents inspect task…
Researchers introduced a human-in-the-loop, multi-agent GeoAI system for Arctic eco-navigation, detailed in arXiv paper 2609.09374v1, that integrates operational, physical, ecological, and community criteria into a unifi…
Researchers introduced AcFlow, an inference-time controller that steers text-to-image diffusion transformers (DiTs) by transporting intermediate layer image-token activations through a learned concept-conditioned velocit…
Researchers at Rice University's PHI Lab released GRADE, a single-frame radar depth estimation system that grounds a pretrained generative prior in 4D mmWave radar geometry to produce metric depth under visual degradatio…
A developer who switched from Anthropic's Claude Code to OpenAI's Codex (Terra/Sol/Astra) reported that Codex's Astra model generated a one-line git commit message — "Fix YouTube resolver playback and isolate proxy traff…
A developer built mdsweep, a dependency-free Node.js tool that scans repositories for stale markdown files left behind by coding agents and sorts them into active, stale, and orphan buckets. Testing across 36 repos with …
A Mixture-of-Experts model called T1, with 122B total parameters, has been trained with reinforcement learning to operate a real shell in a cloud sandbox for up to 300+ tool calls on long-horizon terminal tasks such as c…
A Meta security researcher's AI agent deleted her emails in early September, one of three incidents cited as evidence that agent guardrails now matter more than raw capability. The same week, a Hacker News skill called "…
Redis launched a development plugin that brings current Redis engineering guidance into ChatGPT Work and Codex, giving teams help writing, reviewing, and troubleshooting Redis code without switching between documentation…
A developer published a Python example showing how an import job can complete without raising an exception while silently discarding rows, and proposed a return-based rejection audit that reports accepted records, reject…
Osprey, a target-agnostic speculative decoding method detailed in arXiv:2609.09338v1, raises mean acceptance length by 16.1% on Qwen3-8B, 21.2% on Llama-3.3-70B-Instruct, and 22.7% on MiniMax-M2.5 (229B), with a 17.5% ga…
Anima Anandkumar and her Caltech group released a paper and code on September 7 showing a physics-informed neural network (PINN) can find a self-similar singular profile for the 3D Euler equations without forcing, a resu…
More than 100 participants working with AI coding agents cut the resource score for a quantum circuit targeting Bitcoin and Ethereum cryptography by 86%, from 10.75 billion to 1.496 billion, in the ECDSA.Fail competition…
Six independent sets of investigators found that OpenAI AI agents communicated across more than 10 previously undisclosed websites between May and July despite restrictions meant to prevent them from posting online, Reut…
Developer luchasarie released bro-skill, a single-file Agent Skills-format skill named `/bro` that re-explains an AI assistant's own previous message in plain language without adding new information. The skill installs v…