# Real-time Indexing for Large-scale Recommendation by Streaming Vector Quantization Retriever

arXiv: https://arxiv.org/abs/2501.08695 · HTML: https://mlsanity.com/paper/2501.08695/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Jan 2025.

## Built on by (most cited first, 9 of 9)

- [Stop Treating Collisions Equally: Qualification-Aware Semantic ID Learning for Recommendation at Industrial Scale](https://mlsanity.com/paper/2603.00632/) (2026)
- [Brownian Bridge Diffusion for Sequential Recommendation](https://mlsanity.com/paper/2507.06121/) (2025)
- [MISS: Multi-Modal Tree Indexing and Searching with Lifelong Sequential Behavior for Retrieval Recommendation](https://mlsanity.com/paper/2508.14515/) (2025)
- [UniGRec: Unified Generative Recommendation with Soft Identifiers for End-to-End Optimization](https://mlsanity.com/paper/2601.17438/) (2026)
- [Hierarchical Group-wise Ranking Framework for Recommendation Models](https://mlsanity.com/paper/2506.12756/) (2025)
- [Bidding-Aware Retrieval for Multi-Stage Consistency in Online Advertising](https://mlsanity.com/paper/2508.05206/) (2025)
- [ScaleToT: Generalizing Structured LLM Reasoning for Billion-Scale Low-Activity User Modeling](https://mlsanity.com/paper/2606.24605/) (2026)
- [OneShot: Index-in-Ranking with Neural Scoring for Large-Scale Retrieval](https://mlsanity.com/paper/2607.27475/) (2026)
- [LIGE-GR: A Smooth Leap from Ranking to Generative Recommendation in the LLM Era](https://mlsanity.com/paper/2609.18148/) (2026)

## Abstract (arXiv)

Retrievers, which form one of the most important recommendation stages, are responsible for efficiently selecting possible positive samples to the later stages under strict latency limitations. Because of this, large-scale systems always rely on approximate calculations and indexes to roughly shrink candidate scale, with a simple ranking model. Considering simple models lack the ability to produce precise predictions, most of the existing methods mainly focus on incorporating complicated ranking models. However, another fundamental problem of index effectiveness remains unresolved, which also bottlenecks complication. In this paper, we propose a novel index structure: streaming Vector Quantization model, as a new generation of retrieval paradigm. Streaming VQ attaches items with indexes in real time, granting it immediacy. Moreover, through meticulous verification of possible variants, it achieves additional benefits like index balancing and reparability, enabling it to support complicated ranking models as existing approaches. As a lightweight and implementation-friendly architecture, streaming VQ has been deployed and replaced all major retrievers in Douyin and Douyin Lite, resulting in remarkable user engagement gain.