DREAM Improves Taobao Feed Metrics Without Replacing Rankers
An intent engine and agentic meta-control layer steer existing recommendation stages, lifting GMV 1.31% when extended through fine ranking.
Aug 15, 20264 min2608.09408
Indexing, dictionaries, retrieval, content and analysis.
An intent engine and agentic meta-control layer steer existing recommendation stages, lifting GMV 1.31% when extended through fine ranking.
ROCS separates request-side and candidate-side computation, reaching up to 3× higher retrieval QPS and 50% higher ranking QPS in production.
A compact knowledge-base backbone queries full user histories at each layer, improving offline ranking metrics while cutting per-example FLOPs by 8×–25×.
A novel alignment objective maps passages across 100+ languages to a shared space, outperforming prior retrievers by 4-8 points.