University of Manchester forecasts UK air pollution with NVIDIA Earth-2 generative models
nordicinst · x · 2026-09-16
University of Manchester researchers trained NVIDIA Earth-2 CorrDiff and StormCast on the Isambard-AI supercomputer to forecast UK air pollution, replacing slow chemistry-based models with generative frameworks — the model worked on the first attempt, with workflows demonstrated on a desktop DGX Spark.
More from Research
- New arXiv paper shows 'memorisation bias': medical AI trained on a patient's history can miss new conditions — chaumian · 2026-09-16
- 27-Year-Old Open Problem Closed: Sharp SVM Margin-Based Generalization Bound Proven — LucaAmb · 2026-09-16
- InceptionRAG: Dormant-Passage Poisoning Attack Hits 80%+ Success Against RAG Defenses — chaumian · 2026-09-16
- RSIAgent improves agents without weight updates, beats GPT-6 Astra on OSWorld 2.0 — Roger_M_Taylor · 2026-09-16
- Is mathematical abstraction just human assembly code? The case that AI can grind through what we can't — Jsevillamol · 2026-09-16
- AI designs drugs in months, but India takes a year to approve trials while Australia and China take weeks — jajoosam · 2026-09-16