# GemiRec: Interest Quantization and Generation for Multi-Interest Recommendation

arXiv: https://arxiv.org/abs/2510.14626 · HTML: https://mlsanity.com/paper/2510.14626/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 53 references that are in the map.
- First posted on arXiv 16 Oct 2025.

## Builds on (most cited first, 23 of 23)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [Multi-Interest Network with Dynamic Routing for Recommendation at Tmall](https://mlsanity.com/paper/1904.08030/) (2019)
- [On the Theoretical Limitations of Embedding-Based Retrieval](https://mlsanity.com/paper/2508.21038/) (2025)
- [Controllable Multi-Interest Framework for Recommendation](https://mlsanity.com/paper/2005.09347/) (2020)
- [Sparse-Interest Network for Sequential Recommendation](https://mlsanity.com/paper/2102.09267/) (2021)
- [Efficiently Leveraging Multi-level User Intent for Session-based Recommendation via Atten-Mixer Network](https://mlsanity.com/paper/2206.12781/) (2022)
- [Stacked Quantizers for Compositional Vector Compression](https://mlsanity.com/paper/1411.2173/) (2014)
- [Decoupled Side Information Fusion for Sequential Recommendation](https://mlsanity.com/paper/2204.11046/) (2022)
- [Learning Multi-granular Quantized Embeddings for Large-Vocab Categorical Features in Recommender Systems](https://mlsanity.com/paper/2002.08530/) (2020)
- [Lipschitz regularized Deep Neural Networks generalize and are adversarially robust](https://mlsanity.com/paper/1808.09540/) (2018)
- [Re4: Learning to Re-contrast, Re-attend, Re-construct for Multi-interest Recommendation](https://mlsanity.com/paper/2208.08011/) (2022)
- [Equivariant Contrastive Learning for Sequential Recommendation](https://mlsanity.com/paper/2211.05290/) (2022)
- [When Multi-Level Meets Multi-Interest: A Multi-Grained Neural Model for Sequential Recommendation](https://mlsanity.com/paper/2205.01286/) (2022)
- [Rethinking Multi-Interest Learning for Candidate Matching in Recommender Systems](https://mlsanity.com/paper/2302.14532/) (2023)
- [Exploring Periodicity and Interactivity in Multi-Interest Framework for Sequential Recommendation](https://mlsanity.com/paper/2106.04415/) (2021)
- [Curse of "Low" Dimensionality in Recommender Systems](https://mlsanity.com/paper/2305.13597/) (2023)
- [Diverse Preference Augmentation with Multiple Domains for Cold-start Recommendations](https://mlsanity.com/paper/2204.00327/) (2022)
- [Improving Micro-video Recommendation via Contrastive Multiple Interests](https://mlsanity.com/paper/2205.09593/) (2022)
- [Attribute Simulation for Item Embedding Enhancement in Multi-interest Recommendation](https://mlsanity.com/paper/2311.17374/) (2023)

## Abstract (arXiv)

Multi-interest recommendation has gained attention, especially in industrial retrieval stage. Unlike classical dual-tower methods, it generates multiple user representations instead of a single one to model comprehensive user interests. However, prior studies have identified two underlying limitations: the first is interest collapse, where multiple representations homogenize. The second is insufficient modeling of interest evolution, as they struggle to capture latent interests absent from a user's historical behavior. We begin with a thorough review of existing works in tackling these limitations. Then, we attempt to tackle these limitations from a new perspective. Specifically, we propose a framework-level refinement for multi-interest recommendation, named GemiRec. The proposed framework leverages interest quantization to enforce a structural interest separation and interest generation to learn the evolving dynamics of user interests explicitly. It comprises three modules: (a) Interest Dictionary Maintenance Module (IDMM) maintains a shared quantized interest dictionary. (b) Multi-Interest Posterior Distribution Module (MIPDM) employs a generative model to capture the distribution of user future interests. (c) Multi-Interest Retrieval Module (MIRM) retrieves items using multiple user-interest representations. Both theoretical and empirical analyses, as well as extensive experiments, demonstrate its advantages and effectiveness. Moreover, it has been deployed in production since March 2025, showing its practical value in industrial applications.