Updated daily398 verified stories
Material AI papers, studies, datasets, evaluations, and scientific advances with direct links to the underlying research.
Every story links to the strongest available evidence: original sources when available, otherwise clearly attributed reporting.
What happened, why it matters, and what to watch — no jargon tax.
When the signal is thin, we publish nothing rather than padding the feed.
A growing stream of verified perspectives for people who need to understand AI without chasing hype.
A new arXiv paper describes PARTAB, a framework that selects relevant table regions before an AI model answers questions, reporting improved results on several table-reasoning benchmarks and reduced reasoning context.
The Decoder reports that a UK-Ukraine defense partnership gives approved British firms access to Ukraine’s manually labeled combat imagery and sensor data for military AI development. The arrangement is described as a secure, Ukraine-controlled collaboration involving drone detection, acoustic sensing and low-power…
A new arXiv paper proposes training language models to evaluate their own judgments without external gold labels, while using validation checks to identify when recursive improvement should stop.
A survey of 46 papers examines how natural-language-processing systems support triage, diagnosis and disposition in emergency departments, while identifying unresolved problems with generalisability, noisy clinical inputs and workflow integration.
A new arXiv paper presents an open-source policy schema and supporting tools intended to let organizations define content rules for generative AI applications and apply them across training, alignment and runtime monitoring.
A preprint describes LUCAID, an agentic multimodal AI system that integrates nine lung-cancer pathology tasks and reports 93.0% concordance with an expert-panel reference standard in prospective clinical validation.
An arXiv paper introduces DRRG, a discrete-diffusion framework that generates radiology reports through iterative masked-token denoising. The authors report stronger results than comparison methods on MIMIC-CXR and CheXpert Plus across several metrics, while using a smaller language-model decoder.
A new arXiv paper proposes VisCache, a no-training framework that reduces visual KV-cache storage for vision-language models while retaining 19% to 28% of the cache. The authors report speedups of up to 2.35 times, but the results have not been independently validated here.
A new technical report introduces WeMM-Embedding, a family of 2B, 4B and 9B models for representing text, images, videos and visual documents in a shared space. The report says the models improve internal WeChat benchmarks and have been deployed across search and recommendation services.
A new arXiv study reports that tuning the blend between text and image prototypes is not the main limit on few-shot vision-language model accuracy. In experiments spanning 4,800 evaluation cells, validation-free linear probes outperformed even a blend selected with test-set information.
A new arXiv study comparing primate vision with multiple neural-network designs reports that predictive world models were most robust when object appearance changed, pointing to progressive integration of motion as a missing principle for dynamic AI.
A new preprint presents NeuralParker, a reinforcement-learning planner designed to guide delivery and service vehicles to specified poses in irregular bounded areas, with reported success in a real-vehicle evaluation.
One useful briefing each week
Get the week’s verified AI news, original data, useful tools, learning picks, and fresh AI jobs.
Hiring an AI professional or launching a useful AI product? Put it in front of people who came here to learn and act.
Post an AI job Submit an AI tool