Microsoft's PBC doubles 3-4 week forecast skill, sweeps ECMWF AI Weather Quest

LesterMackey · x · 2026-10-14

Microsoft Research published a Probabilistic Bias Correction (PBC) framework in Nature that corrects predictable errors in physics-based and AI weather forecasts over the unreliable 2-6 week window.

Key results:

• Doubled the skill of ECMWF's AI forecasting system at 3-4 week lead times
• Placed first in ECMWF's AI Weather Quest for every variable and lead time
• Led by former MSR interns Hannah Guan and Soukayna Mouatadid with Paulo Orenstein, Judah Cohen and others

The lightweight ML method matters because farmers, resource managers and disaster responders must make critical decisions weeks in advance, beyond the 10-day reliability horizon of standard forecasts.

Related event: Microsoft's PBC machine learning bias-correction method published in Nature boosts 2-6 week forecasts(10 posts)→

Original post →

More from Research

• Initialization matters far more than expected in LLM-driven scientific discovery, new study finds — BenBlaiszik · 2026-10-15
• RLVR is just RL — RLHF is the weird one, argues lateinteraction — lateinteraction · 2026-10-15
• Mathematician models OpenAI's 372-theorem drop as a self-organized criticality collapse — littmath · 2026-10-15
• RoPE long-context failures proven inevitable; training-free RoPE Profiler boosts accuracy up to 25% — zhaofeng_wu · 2026-10-15
• OpenAI's Quasi-Riemann proof sparks chain of results, Bhargava proves primitive root theorem — Singularitarian · 2026-10-15
• Applied Compute's Spork post-training recipe doubles chess puzzle solve-rate vs GRPO — ypatil125 · 2026-10-15

Research channel →