AWhat changed
Integrated neural candidate generation using user-history embeddings and approximate retrieval with a separate richer-feature ranking network trained toward expected watch time.
Scientific scope
Integrated neural candidate generation using user-history embeddings and approximate retrieval with a separate richer-feature ranking network trained toward expected watch time. Two-stage retrieval/ranking and embeddings predate the work; contribution is the described neural industrial integration and training lessons. Still needs substantial engineered features; watch time is a proxy, not measured user satisfaction or causal welfare.2016system account is not current YouTube architecture or every proprietary component.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Two-stage retrieval/ranking and embeddings predate the work; contribution is the described neural industrial integration and training lessons. Still needs substantial engineered features; watch time is a proxy, not measured user satisfaction or causal welfare.2016system account is not current YouTube architecture or every proprietary component.
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