AWhat changed
Adapted dynamic routing to aggregate variable-length behavior embeddings into several interest vectors, using label-aware attention in training and independent nearest-neighbor queries for candidate retrieval.
Scientific scope
Adapted dynamic routing to aggregate variable-length behavior embeddings into several interest vectors, using label-aware attention in training and independent nearest-neighbor queries for candidate retrieval. Reuses capsule routing and approximate retrieval, with a shared bilinear map and randomized routing initialization. Number of interests is set by a bounded history-length heuristic, not discovered with a universal optimality certificate. Label-aware attention is removed for serving; matching-stage retrieval is distinct from target-specific ranking.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Reuses capsule routing and approximate retrieval, with a shared bilinear map and randomized routing initialization. Number of interests is set by a bounded history-length heuristic, not discovered with a universal optimality certificate. Label-aware attention is removed for serving; matching-stage retrieval is distinct from target-specific ranking.
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