aiqbench analysis
FlexCast introduces a novel deep learning architecture capable of generating weather forecasts across variable input configurations using a single parameter set, directly addressing the transferability limitations inherent in traditional fixed-channel models. By leveraging metadata-conditioned adapters and masked set fusion within a U-Transformer framework, this approach significantly enhances model adaptability without requiring retraining for specific field subsets. This advancement represents a meaningful step forward in making deep learning weather prediction more robust and efficient for diverse atmospheric data sources.
FlexCast: Adaptive Weather Forecasting from Arbitrary Field Sets
arXiv:2610.05296v1 Announce Type: new Abstract: Most deep learning weather models assign a fixed set of variables and pressure levels to predefined channels, limiting transfer across atmospheric field configurations. This dependence on a fixed field set limits the transferability of trained models across atmospheric field configurations. We propose FlexCast, a field-adaptive weather forecasting model that uses a single set of parameters to produce identity-aligned forecasts for variable-cardinality subsets drawn from a 69-field ERA5 registry. Specifically, a metadata-conditioned adapter the first encodes variable identity, pressure level, and field type and combines them with spatial features. Then, shared rank-16 projec?tions are modulated by metadata-dependent gates to produce field?specific features, while masked set fusion aggregates the available fields into a fixed-width representation. Subsequently, a multiscale U-Transformer processes the fused atmospheric features, while an identity-aware query decoder produces forecasts for the requested fields. Finally, FlexCast learns a standardized six-hour increment and applies it recursively to generate forecasts at longer lead times. Experiments on the 2020 ERA5 test set demonstrate that FlexCast operates across varying field configurations. Compatible cross-field context is associated with lower forecast errors, whereas mismatched context increases them.
Abstract:Most deep learning weather models assign a fixed set of variables and pressure levels to predefined channels, limiting transfer across atmospheric field configurations. This dependence on a fixed field set limits the transferability of trained models across atmospheric field configurations. We propose FlexCast, a field-adaptive weather forecasting model that uses a single set of parameters to produce identity-aligned forecasts for variable-cardinality subsets drawn from a 69-field ERA5 registry. Specifically, a metadata-conditioned adapter the first encodes variable identity, pressure level, and field type and combines them with spatial features. Then, shared rank-16 projec?tions are modulated by metadata-dependent gates to produce field?specific features, while masked set fusion aggregates the available fields into a fixed-width representation. Subsequently, a multiscale U-Transformer processes the fused atmospheric features, while an identity-aware query decoder produces forecasts for the requested fields. Finally, FlexCast learns a standardized six-hour increment and applies it recursively to generate forecasts at longer lead times. Experiments on the 2020 ERA5 test set demonstrate that FlexCast operates across varying field configurations. Compatible cross-field context is associated with lower forecast errors, whereas mismatched context increases them.
Machine Learning (cs.LG)
arXiv:2610.05296 [cs.LG]
(or arXiv:2610.05296v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.05296 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 274 words
Related coverage
ORBIT-2 based Weather and Climate Downscaling and Downscaled Global Forecasts on AMD Instinct
AMD · Jun 8, 2026
2026.6: Pick a card, any card
Home Assistant · Jun 3, 2026
Aurora 1.5: Extending open foundation models for weather and Earth-system applications
Microsoft Research · Jul 9, 2026
UniGIO: Unified Generative Global In-situ Weather Modeling from Spatiotemporal Incomplete Observations
arXiv · Sep 22, 2026