# RQ-GMM: Residual Quantized Gaussian Mixture Model for Multimodal Semantic Discretization in CTR Prediction

arXiv: https://arxiv.org/abs/2602.12593 · HTML: https://mlsanity.com/paper/2602.12593/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 37 references that are in the map.
- First posted on arXiv 13 Feb 2026.

## Builds on (most cited first, 24 of 24)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Deep Interest Network for Click-Through Rate Prediction](https://mlsanity.com/paper/1706.06978/) (2017)
- [Deep & Cross Network for Ad Click Predictions](https://mlsanity.com/paper/1708.05123/) (2017)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment](https://mlsanity.com/paper/2502.18965/) (2025)
- [Product-based Neural Networks for User Response Prediction](https://mlsanity.com/paper/1611.00144/) (2016)
- [Deep Learning over Multi-field Categorical Data: A Case Study on User Response Prediction](https://mlsanity.com/paper/1601.02376/) (2016)
- [Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations](https://mlsanity.com/paper/2306.08121/) (2023)
- [Towards Understanding the Overfitting Phenomenon of Deep Click-Through Rate Prediction Models](https://mlsanity.com/paper/2209.06053/) (2022)
- [Personalized Graph Signal Processing for Collaborative Filtering](https://mlsanity.com/paper/2302.02113/) (2023)
- [AgentCF++: Memory-enhanced LLM-based Agents for Popularity-aware Cross-domain Recommendations](https://mlsanity.com/paper/2502.13843/) (2025)
- [Enhancing Taobao Display Advertising with Multimodal Representations: Challenges, Approaches and Insights](https://mlsanity.com/paper/2407.19467/) (2024)
- [AutoSeqRec: Autoencoder for Efficient Sequential Recommendation](https://mlsanity.com/paper/2308.06878/) (2023)
- [Parameter-free Dynamic Graph Embedding for Link Prediction](https://mlsanity.com/paper/2210.08189/) (2022)
- [Triple Structural Information Modelling for Accurate, Explainable and Interactive Recommendation](https://mlsanity.com/paper/2304.11528/) (2023)
- [Improving LLM-powered Recommendations with Personalized Information](https://mlsanity.com/paper/2502.13845/) (2025)
- [Recommendation Unlearning via Matrix Correction](https://mlsanity.com/paper/2307.15960/) (2023)
- [Unbiased Collaborative Filtering with Fair Sampling](https://mlsanity.com/paper/2502.13840/) (2025)
- [Filtering Discomforting Recommendations with Large Language Models](https://mlsanity.com/paper/2410.05411/) (2024)
- [FedCIA: Federated Collaborative Information Aggregation for Privacy-Preserving Recommendation](https://mlsanity.com/paper/2504.14208/) (2025)
- [Dynamic Feature-Embedding Communication via Codebook Distillation for Federated Recommendation](https://mlsanity.com/paper/2601.18570/) (2026)

## Built on by (most cited first, 1 of 1)

- [PRQ-KMeans: Projection Residual Quantization for Semantic ID Tokenization](https://mlsanity.com/paper/2608.24207/) (2026)

## Abstract (arXiv)

Multimodal content is crucial for click-through rate (CTR) prediction. However, directly incorporating continuous embeddings from pre-trained models into CTR models yields suboptimal results due to misaligned optimization objectives and convergence speed inconsistency during joint training. Discretizing embeddings into semantic IDs before feeding them into CTR models offers a more effective solution, yet existing methods suffer from limited codebook utilization, reconstruction accuracy, and semantic discriminability. We propose RQ-GMM (Residual Quantized Gaussian Mixture Model), which introduces probabilistic modeling to better capture the statistical structure of multimodal embedding spaces. Through Gaussian Mixture Models combined with residual quantization, RQ-GMM achieves superior codebook utilization and reconstruction accuracy. Experiments on public datasets and online A/B tests on a large-scale short-video platform serving hundreds of millions of users demonstrate substantial improvements: RQ-GMM yields a 1.502% gain in Advertiser Value over strong baselines. The method has been fully deployed, serving daily recommendations for hundreds of millions of users.