Unified retrieval and ranking system improves mobile feed recommendations
UNIQUE: A Unified Retrieval and Ranking System for Large-Scale Feed Recommendation
Information Retrieval
Summary
Serving content on mobile feeds requires quickly finding and ranking lots of different items, but existing methods often lose important information and struggle with less popular or new content. The authors created UNIQUE, which combines retrieval and ranking into a single system using a special way to compress and represent content, leading to better recommendations overall. Their system achieves more balanced treatment of all content and performs well in real-world tests, improving watch time and distribution while keeping fast response times.
What this means in practice
- •For mobile app developers: Improve recommendation quality and speed in mobile feed apps by integrating retrieval and ranking with balanced quantization for better long-tail content exposure.
- •For video platform engineers: Enhance short-video content recommendation by using a unified model that boosts watch duration and manages diverse, fast-changing video feeds efficiently.
Authors
Zhuang Liu, Yongkang Fu, Zuodong Yang, Guangxing Chen, Zonggang Wu, Yuqi Lu, Shouke Qin, Shantao Li, Maolin Wang
Abstract
Industrial mobile feed systems rely on a retrieval-ranking pipeline to serve large-scale, heterogeneous, and fast-changing content under strict latency constraints. However, existing pipelines still suffer from two critical issues: hierarchical quantization instability in candidate retrieval and information loss between separated retrieval and ranking stages. These issues hurt long-tail and cold-start recommendation and complicate efficient serving. To address them, we present UNIQUE, a unified retrieval and ranking recommendation framework with single-layer flat quantization. UNIQUE integrates generative code-based retrieval and target-aware ranking into one early-fusion architecture, enabling end-to-end training under a shared representation while preserving efficient candidate generation. A balanced quantization mechanism is further introduced to mitigate codebook imbalance and improve long-tail representation. Offline experiments evaluate UNIQUE from both retrieval and ranking perspectives, while codebook analysis shows more balanced resource allocation than hierarchical quantization. We deploy UNIQUE in the homepage feed, discovery-page, and short-video recommendation scenarios of Mobile Baidu, serving large-scale real-world traffic. Online A/B tests achieve a 0.96% gain in total watch duration and a 1.08% gain in total distribution volume, with notable improvements for new users and highly active users. Serving measurements show 89 ms P99 latency and 44.23% online inference MFU. These results show that UNIQUE provides a stable, efficient, and production-ready framework for unified retrieval and ranking in industrial recommendation.