# kNN-Embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrieval

arXiv: https://arxiv.org/abs/2205.06205 · HTML: https://mlsanity.com/paper/2205.06205/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 40 references that are in the map.
- First posted on arXiv 12 May 2022.

## Builds on (most cited first, 16 of 16)

- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https://mlsanity.com/paper/1806.01973/) (2018)
- [Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphs](https://mlsanity.com/paper/1603.09320/) (2016)
- [Generalization through Memorization: Nearest Neighbor Language Models](https://mlsanity.com/paper/1911.00172/) (2019)
- [Nearest Neighbor Machine Translation](https://mlsanity.com/paper/2010.00710/) (2020)
- [Embedding-based Retrieval in Facebook Search](https://mlsanity.com/paper/2006.11632/) (2020)
- [PyTorch-BigGraph: A Large-scale Graph Embedding System](https://mlsanity.com/paper/1903.12287/) (2019)
- [word2vec Explained: deriving Mikolov et al.'s negative-sampling word-embedding method](https://mlsanity.com/paper/1402.3722/) (2014)
- [Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS)](https://mlsanity.com/paper/1405.5869/) (2014)
- [PinnerSage: Multi-Modal User Embedding Framework for Recommendations at Pinterest](https://mlsanity.com/paper/2007.03634/) (2020)
- [Unbounded cache model for online language modeling with open vocabulary](https://mlsanity.com/paper/1711.02604/) (2017)
- [TwHIN: Embedding the Twitter Heterogeneous Information Network for Personalized Recommendation](https://mlsanity.com/paper/2202.05387/) (2022)
- [Related Pins at Pinterest: The Evolution of a Real-World Recommender System](https://mlsanity.com/paper/1702.07969/) (2017)
- [Candidate Generation with Binary Codes for Large-Scale Top-N Recommendation](https://mlsanity.com/paper/1909.05475/) (2019)
- [Massively Multilingual Document Alignment with Cross-lingual Sentence-Mover's Distance](https://mlsanity.com/paper/2002.00761/) (2020)

## Built on by (most cited first, 4 of 4)

- [TwHIN-BERT: A Socially-Enriched Pre-trained Language Model for Multilingual Tweet Representations at Twitter](https://mlsanity.com/paper/2209.07562/) (2022)
- [NTULM: Enriching Social Media Text Representations with Non-Textual Units](https://mlsanity.com/paper/2210.16586/) (2022)
- [MUSS: Multilevel Subset Selection for Relevance and Diversity](https://mlsanity.com/paper/2503.11126/) (2025)
- [URL-BERT: Training Webpage Representations via Social Media Engagements](https://mlsanity.com/paper/2310.16303/) (2023)

## Abstract (arXiv)

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 "interests" of the user. By querying each of a user's mixture component in proportion to their mixture weights, we retrieve a high-diversity set of candidates reflecting elements from each of a user's interests. We experimentally compare kNN-Embed to standard ANN candidate retrieval, and show significant improvements in overall recall and improved diversity across three datasets. Accompanying this work, we open source a large Twitter follow-graph dataset (https://huggingface.co/datasets/Twitter/TwitterFollowGraph), to spur further research in graph-mining and representation learning for recommender systems.