# Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations

arXiv: https://arxiv.org/abs/2306.08121 · HTML: https://mlsanity.com/paper/2306.08121/

## Key facts

- Cited by 52 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 36 references that are in the map.
- First posted on arXiv 13 Jun 2023.

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

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Scaling Autoregressive Models for Content-Rich Text-to-Image Generation](https://mlsanity.com/paper/2206.10789/) (2022)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https://mlsanity.com/paper/1806.01973/) (2018)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Wide & Deep Learning for Recommender Systems](https://mlsanity.com/paper/1606.07792/) (2016)
- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [VideoBERT: A Joint Model for Video and Language Representation Learning](https://mlsanity.com/paper/1904.01766/) (2019)
- [Vector-quantized Image Modeling with Improved VQGAN](https://mlsanity.com/paper/2110.04627/) (2021)
- [Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates](https://mlsanity.com/paper/1804.10959/) (2018)
- [Jukebox: A Generative Model for Music](https://mlsanity.com/paper/2005.00341/) (2020)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [Where to Go Next for Recommender Systems? ID- vs. Modality-based Recommender Models Revisited](https://mlsanity.com/paper/2303.13835/) (2023)
- [Neural Collaborative Filtering vs. Matrix Factorization Revisited](https://mlsanity.com/paper/2005.09683/) (2020)
- [Learning Vector-Quantized Item Representation for Transferable Sequential Recommenders](https://mlsanity.com/paper/2210.12316/) (2022)
- [PinnerFormer: Sequence Modeling for User Representation at Pinterest](https://mlsanity.com/paper/2205.04507/) (2022)
- [Learning to Embed Categorical Features without Embedding Tables for Recommendation](https://mlsanity.com/paper/2010.10784/) (2020)
- [Model Size Reduction Using Frequency Based Double Hashing for Recommender Systems](https://mlsanity.com/paper/2007.14523/) (2020)
- [TransRec: Learning Transferable Recommendation from Mixture-of-Modality Feedback](https://mlsanity.com/paper/2206.06190/) (2022)
- [Learning Multi-granular Quantized Embeddings for Large-Vocab Categorical Features in Recommender Systems](https://mlsanity.com/paper/2002.08530/) (2020)
- [Improving Training Stability for Multitask Ranking Models in Recommender Systems](https://mlsanity.com/paper/2302.09178/) (2023)

## Built on by (most cited first, 30 of 52)

- [Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models](https://mlsanity.com/paper/2601.07372/) (2026)
- [Sparse Meets Dense: Unified Generative Recommendations with Cascaded Sparse-Dense Representations](https://mlsanity.com/paper/2503.02453/) (2025)
- [PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations](https://mlsanity.com/paper/2510.07784/) (2025)
- [CALRec: Contrastive Alignment of Generative LLMs for Sequential Recommendation](https://mlsanity.com/paper/2405.02429/) (2024)
- [Enhancing Taobao Display Advertising with Multimodal Representations: Challenges, Approaches and Insights](https://mlsanity.com/paper/2407.19467/) (2024)
- [Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID](https://mlsanity.com/paper/2504.02137/) (2025)
- [Unifying Generative and Dense Retrieval for Sequential Recommendation](https://mlsanity.com/paper/2411.18814/) (2024)
- [ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation](https://mlsanity.com/paper/2502.13581/) (2025)
- [Time to Split: Exploring Data Splitting Strategies for Offline Evaluation of Sequential Recommenders](https://mlsanity.com/paper/2507.16289/) (2025)
- [Embedding-to-Prefix: Parameter-Efficient Personalization for Pre-Trained Large Language Models](https://mlsanity.com/paper/2505.17051/) (2025)
- [MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation](https://mlsanity.com/paper/2508.15281/) (2025)
- [STAR: A Simple Training-free Approach for Recommendations using Large Language Models](https://mlsanity.com/paper/2410.16458/) (2024)
- [Beyond Retrieval: Generating Narratives in Conversational Recommender Systems](https://mlsanity.com/paper/2410.16780/) (2024)
- [PinRec: Unified Generative Retrieval for Pinterest Recommender Systems](https://mlsanity.com/paper/2504.10507/) (2025)
- [Item-Language Model for Conversational Recommendation](https://mlsanity.com/paper/2406.02844/) (2024)
- [Masked Diffusion for Generative Recommendation](https://mlsanity.com/paper/2511.23021/) (2025)
- [Preference Discerning with LLM-Enhanced Generative Retrieval](https://mlsanity.com/paper/2412.08604/) (2024)
- [Stop Treating Collisions Equally: Qualification-Aware Semantic ID Learning for Recommendation at Industrial Scale](https://mlsanity.com/paper/2603.00632/) (2026)
- [IGD: Token Decisiveness Modeling via Information Gain in LLMs for Personalized Recommendation](https://mlsanity.com/paper/2506.13229/) (2025)
- [Pre-train, Align, and Disentangle: Empowering Sequential Recommendation with Large Language Models](https://mlsanity.com/paper/2412.04107/) (2024)
- [A Simple Contrastive Framework Of Item Tokenization For Generative Recommendation](https://mlsanity.com/paper/2506.16683/) (2025)
- [Act-With-Think: Chunk Auto-Regressive Modeling for Generative Recommendation](https://mlsanity.com/paper/2506.23643/) (2025)
- [Towards Scalable Semantic Representation for Recommendation](https://mlsanity.com/paper/2410.09560/) (2024)
- [SIDE: Semantic ID Embedding for effective learning from sequences](https://mlsanity.com/paper/2506.16698/) (2025)
- [Pctx: Tokenizing Personalized Context for Generative Recommendation](https://mlsanity.com/paper/2510.21276/) (2025)
- [Describe What You See with Multimodal Large Language Models to Enhance Video Recommendations](https://mlsanity.com/paper/2508.09789/) (2025)
- [DenseRec: Revisiting Dense Content Embeddings for Sequential Transformer-based Recommendation](https://mlsanity.com/paper/2508.18442/) (2025)
- [Taming the One-Epoch Phenomenon in Online Recommendation System by Two-stage Contrastive ID Pre-training](https://mlsanity.com/paper/2508.18700/) (2025)
- [Sequential Data Augmentation for Generative Recommendation](https://mlsanity.com/paper/2509.13648/) (2025)
- [Language Models as Semantic Augmenters for Sequential Recommenders](https://mlsanity.com/paper/2510.18046/) (2025)

## Abstract (arXiv)

Randomly-hashed item ids are used ubiquitously in recommendation models. However, the learned representations from random hashing prevents generalization across similar items, causing problems of learning unseen and long-tail items, especially when item corpus is large, power-law distributed, and evolving dynamically. In this paper, we propose using content-derived features as a replacement for random ids. We show that simply replacing ID features with content-based embeddings can cause a drop in quality due to reduced memorization capability. To strike a good balance of memorization and generalization, we propose to use Semantic IDs -- a compact discrete item representation learned from frozen content embeddings using RQ-VAE that captures the hierarchy of concepts in items -- as a replacement for random item ids. Similar to content embeddings, the compactness of Semantic IDs poses a problem of easy adaption in recommendation models. We propose novel methods for adapting Semantic IDs in industry-scale ranking models, through hashing sub-pieces of of the Semantic-ID sequences. In particular, we find that the SentencePiece model that is commonly used in LLM tokenization outperforms manually crafted pieces such as N-grams. To the end, we evaluate our approaches in a real-world ranking model for YouTube recommendations. Our experiments demonstrate that Semantic IDs can replace the direct use of video IDs by improving the generalization ability on new and long-tail item slices without sacrificing overall model quality.