Semantic IDs for Music Recommendation
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_ | 1866912500159610880 |
|---|---|
| author | Mei, M. Jeffrey Henkel, Florian Sandberg, Samuel E. Bembom, Oliver Ehmann, Andreas F. |
| author_facet | Mei, M. Jeffrey Henkel, Florian Sandberg, Samuel E. Bembom, Oliver Ehmann, Andreas F. |
| contents | Training recommender systems for next-item recommendation often requires unique embeddings to be learned for each item, which may take up most of the trainable parameters for a model. Shared embeddings, such as using content information, can reduce the number of distinct embeddings to be stored in memory. This allows for a more lightweight model; correspondingly, model complexity can be increased due to having fewer embeddings to store in memory. We show the benefit of using shared content-based features ('semantic IDs') in improving recommendation accuracy and diversity, while reducing model size, for two music recommendation datasets, including an online A/B test on a music streaming service. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_18800 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Semantic IDs for Music Recommendation Mei, M. Jeffrey Henkel, Florian Sandberg, Samuel E. Bembom, Oliver Ehmann, Andreas F. Information Retrieval Machine Learning Training recommender systems for next-item recommendation often requires unique embeddings to be learned for each item, which may take up most of the trainable parameters for a model. Shared embeddings, such as using content information, can reduce the number of distinct embeddings to be stored in memory. This allows for a more lightweight model; correspondingly, model complexity can be increased due to having fewer embeddings to store in memory. We show the benefit of using shared content-based features ('semantic IDs') in improving recommendation accuracy and diversity, while reducing model size, for two music recommendation datasets, including an online A/B test on a music streaming service. |
| title | Semantic IDs for Music Recommendation |
| topic | Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2507.18800 |