Ai2's HiRO-ACEv2 simulates global extreme weather at 3 km for ~1/300th the cost of X-SHiELD
allen_ai · x · 2026-10-15
Allen AI (Ai2) unveils HiRO-ACEv2, which simulates global extreme weather at 3 km resolution for roughly 1/300th the cost of X-SHiELD, matching its statistics within 3% even for events as rare as 1 in a million data points.
The system works in two stages: ACE, an AI climate emulator, simulates climate at a coarser 100 km resolution; HiROv2, an AI downscaling model, sharpens outputs to 3 km. Version 2 adds downscaling of near-surface winds and sea level pressure (v1 handled rainfall only). Savings are even larger for individual regions.
More from Research
- Random Brainfuck programs on a 2D grid spontaneously evolve self-replicators — spikedoanz · 2026-10-15
- Classifier run on 42,000+ ICLR 2027 submissions finds over half nearly fully AI-assisted — nsaphra · 2026-10-15
- LLM4Code workshop at ICSE 2027 opens call for papers, due Nov 13, 2026 — LingmingZhang · 2026-10-15
- ELSA3D, a unified 3D generation and captioning model, hits SOTA at half the FLOPs, accepted at NeurIPS — CSProfKGD · 2026-10-15
- Moving ViT weights 1.3% without changing predictions scrambles representations, challenging Platonic hypothesis — neuroecology · 2026-10-15
- NeurIPS Spotlight: time-reversed LMs (TRLM) boost best-of-N reranking by up to 5% — LucaAmb · 2026-10-15