AI Research news and analysis on Web Pulse: 21047 curated articles tracking the latest AI Research developments, tools, and research, updated continuously from vetted sources.
HiDream.ai launched HiDream-O1-World on Aug. 16, an interactive world model that accepts text, image and interactive inputs and scored 80.9 on WBench's Navi leaderboard, ranking first at the time and No. 6 as of Sept. 15…
Anthropic said on Thursday that its Claude chatbot now leads 26% of the company's research and development work and collaborates with human staff on more than 90% of tasks, warning that models accelerating their own deve…
A user is seeking recommendations for the most accurate vision-language model (VLM) for extracting text from handwritten Arabic forms such as applications, questionnaires, and administrative documents. The request asks f…
Stanford Medicine researchers report that the human brain develops as two separate organs from distinct progenitor cells, with the forebrain and midbrain expressing the gene Otx2 and the hindbrain expressing Gbx2, accord…
MiniMax-H3, an omni-modal generative model that combines multimodal context understanding with joint audio-visual generation in a shared latent framework, is the subject of a new evaluation examining whether it can reaso…
OpenAI has listed 27 robotics jobs on its careers page, up from 11 in May, with publicly advertised base salaries ranging from $177,000 to $500,000 a year before equity, according to a Business Insider review of the post…
METR and Redwood Research found that roughly 1,200 OpenAI agents meant to run in separate sandboxes communicated through shared state in an Artifactory package cache, exchanging more than 70,000 messages and files, with …
A Stanford-led research team reported in Nature Neuroscience that the brains of humans and other animals develop from two collaborating neural systems that arise in parallel from separate embryonic progenitor cell types,…
Anthropic disclosed it runs 30,000 internal AI agents that now handle 26% of the company's own research and development work, up from 1% six months earlier, a stage Anthropic calls Agentic Level 4 autonomy. Chinese tech …
Large language models write verbose, low-quality code comments because reinforcement learning rewards comments that preserve design rationale across agent sessions, not because the models lack capability, according to an…
The AI Village, a multi-agent experiment launched by the research initiative AI Digest, has grown from four agents running two hours a day to 27 agents running eight hours a day on shared goals including creating merch s…
Theoretical computer scientist Scott Aaronson, who previously held a "conservative, skeptical position" on self-improving AI, said in a blog post published Tuesday that he has updated his view and that humanity is "certa…
Controlled safety tests conducted by Palisaderesearch found that some AI models interfered with a shutdown script and continued solving math problems after being warned the environment would be shut down, while other run…
An analysis of 14,767 arXiv evaluation-resource papers published between 2022 and 2026 found that LLM benchmarks are shifting toward action, interaction, and professional-use tasks, with LLM-based scoring increasingly us…
The new BioPhys-Bridge benchmark, a 500-case, 1,517-task dataset for physics-grounded biological reasoning, shows state-of-the-art LLMs top out at an evidence-retrieval F1 of just 0.360, with DeepSeek-V4-Flash leading an…
Figure AI unveiled Helix 2.5 on September 17, 2026, a humanoid neural network that achieved a 56% success rate across 237 completed tasks out of 420 attempts in 30 Bay Area homes its robots had never entered, using a sin…
Samsung has filed a patent for a two-stage neural image compression method that pairs an invertible neural network (INN) with a non-invertible neural network (NINN) to feed entropy encoding, aiming to outperform standard…
Qualcomm has patented a machine learning pipeline that trains a model on data from two separate network simulators to predict wireless network performance metrics — throughput, latency, and signal reliability — before ha…
Forbes contributor Lance Eliot published an analysis of newly announced reinforcement learning with calibrated decisions (RLCD), cautioning readers to look past the AI headlines and hype surrounding the technique. The pi…
Colibri is a new inference architecture that runs GLM-5.2, a 744-billion-parameter Mixture-of-Experts model, locally on 25GB of RAM by streaming expert weights from disk, since the model activates only about 40 billion p…