Space, science, new instruments. A place to discover developments quickly, then stay with the questions that matter to you.
Newest source publications first. These are source dates, not live activity.
Google DeepMind reported that AlphaEvolve improved a trained graph neural network's ability to find feasible solutions to the AC Optimal Power Flow problem from 14% to more than 88%.
Demonstrated · Google DeepMind / Google ResearchAlphaGenome is a unified DNA-sequence model that takes up to 1 Mb of DNA and predicts thousands of functional genomic tracks, including gene expression, splicing, chromatin features, transcription-factor binding, and contact maps.
Demonstrated · Google DeepMindGoogle DeepMind and Commonwealth Fusion Systems announced a partnership in which CFS uses the open-source TORAX plasma simulator in daily workflows to test and refine SPARC operating plans before the tokamak is turned on.
Emerging · Google DeepMind / Commonwealth Fusion SystemsResearchers developed Deep Loop Shaping, a reinforcement-learning feedback controller tested on the LIGO Livingston Observatory to reduce control noise in a difficult mirror-control loop.
Demonstrated · Google DeepMind / LIGO / Caltech / Gran Sasso Science InstituteA multicenter randomized, controlled, double-blind superiority trial tested whether adding AI-detected spatio-temporal electrogram-dispersion targets to pulmonary-vein isolation improves treatment of drug-refractory persistent atrial fibrillation.
Demonstrated · TAILORED-AF InvestigatorsFunSearch combined language models with program evaluation to discover mathematical constructions.
Demonstrated · Google DeepMindGNoME used graph networks to search for energetically favorable crystal structures.
Experimental · Google DeepMindScientists used machine learning to help engineer an enzyme that breaks down PET plastic.
Experimental · University of Texas at AustinHistorians working with Ithaca improved the restoration of damaged ancient Greek inscriptions.
Demonstrated · Ca’ Foscari University & DeepMindResearchers used reinforcement learning to control magnetic confinement in a research tokamak.
Demonstrated · EPFL & DeepMindStarcloud says its Starcloud-1 satellite launched with an NVIDIA H100 and later trained nanoGPT in space.
Emerging · Starcloud / NVIDIA