Method evidence record
FiLM
FiLM is a long-term time-series forecasting architecture introduced by Tian Zhou and colleagues at NeurIPS 2022. It combines Legendre polynomial projections of the historical input with learnable frequency-domain filters applied to the resulting coefficient sequences. By representing history as a compact set of polynomial coefficients and filtering those coefficients in the frequency domain, FiLM enables efficient extrapolation over long prediction horizons without the quadratic cost of full self-attention.
Evidence statusSources recorded, not reviewedBibliographic sources are present. Claim-level evidence review has not been performed.
Source record
Recorded
Citations copied verbatim from the method’s source record. No claim-level verification is inferred from them.
FiLM (Frequency Improved Legendre Memory Model)
Taxonomic method recordml-modeldeep-learning
Open full method1 recorded citation, copied from the method source record.
Curated claims
Assessed
Claims persisted in the evidence ledger, each with its own assessment.
No curated claims yet
This view does not invent a claim assessment when the ledger has none.
Related methods
Suggested3
Generated from the method graph and shown as machine-suggested relations. No evidence claim is inferred.
- Taxonomic bucketAutoformer
machine-suggestedRelational suggestion, not evidence.
- Taxonomic bucketFEDformer
machine-suggestedRelational suggestion, not evidence.
- See alsoState Space Model
machine-suggestedRelational suggestion, not evidence.