Knn-embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrieval

Abstract

Candidate retrieval is the first stage in recommendation systems, where a light-weight system is used to retrieve potentially relevant items for an input user. These candidate items are then ranked and pruned in later stages of recommender systems using a more complex ranking model. As the top of the recommendation funnel, it is important to retrieve a high-recall candidate set to feed into downstream ranking models. A common approach is to leverage approximate nearest neighbor (ANN) search from a single dense query embedding; however, this approach this can yield a low-diversity result set with many near duplicates. As users often have multiple interests, candidate retrieval should ideally return a diverse set of candidates reflective of the user's multiple interests. To this end, we introduce kNN-Embed, a general approach to improving diversity in dense ANN-based retrieval. kNN-Embed represents each user as a smoothed mixture over learned item clusters that represent distinct "intere

Stats

• citations2
• S2 citations—
• github stars0
• HF likes0
• heat score3.58
• arxiv keyelkishky2022knn

Related papers

• Everyone's Preference Changes Differently: Weighted Multi-interest Retrieval Model (2022)0.00
• Adaptive Retrieval And Scalable Indexing For K-nn Search With Cross-encoders (2024)0.00
• Efficient K-nn Search With Cross-encoders Using Adaptive Multi-round CUR Decomposition (2023)0.00
• Synergizing Implicit And Explicit User Interests: A Multi-embedding Retrieval Framework At Pinterest (2025)0.00
• NUDGE: Lightweight Non-parametric Fine-tuning Of Embeddings For Retrieval (2024)0.00
• A Novel User Representation Paradigm For Making Personalized Candidate Retrieval (2019)0.00
• Comparing Neighbors Together Makes It Easy: Jointly Comparing Multiple Candidates For Efficient And Effective Retrieval (2024)4.52
• Deep Retrieval: Learning A Retrievable Structure For Large-scale Recommendations (2020)0.00