AWhat changed
Uses variable-specific encoders/decoders and cross-variable fusion, learns uncertainty-weighted weather losses, and reuses predicted intermediate states from a replay buffer to train long-lead autoregressive forecasts.
Scientific scope
Uses variable-specific encoders/decoders and cross-variable fusion, learns uncertainty-weighted weather losses, and reuses predicted intermediate states from a replay buffer to train long-lead autoregressive forecasts. First four authors share contribution, Lei Bai project lead. Forecast variables treated as modalities, not an image/text multimodal system. Gaussian independent-target loss and replay training do not certify calibrated ensemble uncertainty or eliminate error accumulation. Skill threshold is source-specific, not every variable reliable beyond ten days; ERA5 costs remain.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. First four authors share contribution, Lei Bai project lead. Forecast variables treated as modalities, not an image/text multimodal system. Gaussian independent-target loss and replay training do not certify calibrated ensemble uncertainty or eliminate error accumulation. Skill threshold is source-specific, not every variable reliable beyond ten days; ERA5 costs remain.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction