Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913774677524480 |
|---|---|
| author | Zheng, Carolina Huang, Minhui Pedchenko, Dmitrii Rangadurai, Kaushik Wang, Siyu Nahum, Gaby Lei, Jie Yang, Yang Liu, Tao Luo, Zutian Wei, Xiaohan Ramasamy, Dinesh Yang, Jiyan Han, Yiping Yang, Lin Xu, Hangjun Jin, Rong Yang, Shuang |
| author_facet | Zheng, Carolina Huang, Minhui Pedchenko, Dmitrii Rangadurai, Kaushik Wang, Siyu Nahum, Gaby Lei, Jie Yang, Yang Liu, Tao Luo, Zutian Wei, Xiaohan Ramasamy, Dinesh Yang, Jiyan Han, Yiping Yang, Lin Xu, Hangjun Jin, Rong Yang, Shuang |
| contents | The exponential growth of online content has posed significant challenges to ID-based models in industrial recommendation systems, ranging from extremely high cardinality and dynamically growing ID space, to highly skewed engagement distributions, to prediction instability as a result of natural id life cycles (e.g, the birth of new IDs and retirement of old IDs). To address these issues, many systems rely on random hashing to handle the id space and control the corresponding model parameters (i.e embedding table). However, this approach introduces data pollution from multiple ids sharing the same embedding, leading to degraded model performance and embedding representation instability.
This paper examines these challenges and introduces Semantic ID prefix ngram, a novel token parameterization technique that significantly improves the performance of the original Semantic ID. Semantic ID prefix ngram creates semantically meaningful collisions by hierarchically clustering items based on their content embeddings, as opposed to random assignments. Through extensive experimentation, we demonstrate that Semantic ID prefix ngram not only addresses embedding instability but also significantly improves tail id modeling, reduces overfitting, and mitigates representation shifts. We further highlight the advantages of Semantic ID prefix ngram in attention-based models that contextualize user histories, showing substantial performance improvements. We also report our experience of integrating Semantic ID into Meta production Ads Ranking system, leading to notable performance gains and enhanced prediction stability in live deployments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_02137 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID Zheng, Carolina Huang, Minhui Pedchenko, Dmitrii Rangadurai, Kaushik Wang, Siyu Nahum, Gaby Lei, Jie Yang, Yang Liu, Tao Luo, Zutian Wei, Xiaohan Ramasamy, Dinesh Yang, Jiyan Han, Yiping Yang, Lin Xu, Hangjun Jin, Rong Yang, Shuang Information Retrieval Artificial Intelligence The exponential growth of online content has posed significant challenges to ID-based models in industrial recommendation systems, ranging from extremely high cardinality and dynamically growing ID space, to highly skewed engagement distributions, to prediction instability as a result of natural id life cycles (e.g, the birth of new IDs and retirement of old IDs). To address these issues, many systems rely on random hashing to handle the id space and control the corresponding model parameters (i.e embedding table). However, this approach introduces data pollution from multiple ids sharing the same embedding, leading to degraded model performance and embedding representation instability. This paper examines these challenges and introduces Semantic ID prefix ngram, a novel token parameterization technique that significantly improves the performance of the original Semantic ID. Semantic ID prefix ngram creates semantically meaningful collisions by hierarchically clustering items based on their content embeddings, as opposed to random assignments. Through extensive experimentation, we demonstrate that Semantic ID prefix ngram not only addresses embedding instability but also significantly improves tail id modeling, reduces overfitting, and mitigates representation shifts. We further highlight the advantages of Semantic ID prefix ngram in attention-based models that contextualize user histories, showing substantial performance improvements. We also report our experience of integrating Semantic ID into Meta production Ads Ranking system, leading to notable performance gains and enhanced prediction stability in live deployments. |
| title | Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID |
| topic | Information Retrieval Artificial Intelligence |
| url | https://arxiv.org/abs/2504.02137 |