Semantic IDs for Recommender Systems at Snapchat: Use Cases, Technical Challenges, and Design Choices

2026-04-04

AI Summary

Traditional atomic IDs in recommender systems suffer from high cardinality and lack of semantic meaning, which limits model generalization and personalization. This work proposes Semantic IDs (SIDs)—low-cardinality, semantically clustered, ordered encoding sequences—as a novel representation for recommendation. SIDs are constructed via residual quantization, foundation model embeddings, or collaborative signal extraction, and are integrated into both ranking model features and retrieval sources. By effectively combining semantic information with user behavioral signals, the proposed approach achieves significant offline improvements on both internal datasets and public benchmarks. Multiple SID variants have been deployed in industrial-scale recommender systems, where A/B tests consistently demonstrate stable and positive online gains.

Abstract

Effective item identifiers (IDs) are an important component for recommender systems (RecSys) in practice, and are commonly adopted in many use cases such as retrieval and ranking. IDs can encode collaborative filtering signals within training data, such that RecSys models can extrapolate during the inference and personalize the prediction based on users' behavioral histories. Recently, Semantic IDs (SIDs) have become a trending paradigm for RecSys. In comparison to the conventional atomic ID, an SID is an ordered list of codes, derived from tokenizers such as residual quantization, applied to semantic representations commonly extracted from foundation models or collaborative signals. SIDs have drastically smaller cardinality than the atomic counterpart, and induce semantic clustering in the ID space. At Snapchat, we apply SIDs as auxiliary features for ranking models, and also explore SIDs as additional retrieval sources in different ML applications. In this paper, we discuss practical technical challenges we encountered while applying SIDs, experiments we have conducted, and design choices we have iterated to mitigate these challenges. Backed by promising offline results on both internal data and academic benchmarks as well as online A/B studies, SID variants have been launched in multiple production models with positive metrics impact.

Problem

• Semantic IDs
• Recommender Systems
• Item Identifiers
• Collaborative Filtering
• Retrieval and Ranking

Innovation

• Semantic IDs
• recommender systems
• residual quantization
• foundation models
• semantic clustering