DashCLIP: Leveraging multimodal models for generating semantic embeddings for DoorDash
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914140905275392 |
|---|---|
| author | Gurjar, Omkar Liu, Kin Sum Kolli, Praveen Kumar, Utsaw Rahurkar, Mandar |
| author_facet | Gurjar, Omkar Liu, Kin Sum Kolli, Praveen Kumar, Utsaw Rahurkar, Mandar |
| contents | Despite the success of vision-language models in various generative tasks, obtaining high-quality semantic representations for products and user intents is still challenging due to the inability of off-the-shelf models to capture nuanced relationships between the entities. In this paper, we introduce a joint training framework for product and user queries by aligning uni-modal and multi-modal encoders through contrastive learning on image-text data. Our novel approach trains a query encoder with an LLM-curated relevance dataset, eliminating the reliance on engagement history. These embeddings demonstrate strong generalization capabilities and improve performance across applications, including product categorization and relevance prediction. For personalized ads recommendation, a significant uplift in the click-through rate and conversion rate after the deployment further confirms the impact on key business metrics. We believe that the flexibility of our framework makes it a promising solution toward enriching the user experience across the e-commerce landscape. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_07110 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DashCLIP: Leveraging multimodal models for generating semantic embeddings for DoorDash Gurjar, Omkar Liu, Kin Sum Kolli, Praveen Kumar, Utsaw Rahurkar, Mandar Information Retrieval Machine Learning Despite the success of vision-language models in various generative tasks, obtaining high-quality semantic representations for products and user intents is still challenging due to the inability of off-the-shelf models to capture nuanced relationships between the entities. In this paper, we introduce a joint training framework for product and user queries by aligning uni-modal and multi-modal encoders through contrastive learning on image-text data. Our novel approach trains a query encoder with an LLM-curated relevance dataset, eliminating the reliance on engagement history. These embeddings demonstrate strong generalization capabilities and improve performance across applications, including product categorization and relevance prediction. For personalized ads recommendation, a significant uplift in the click-through rate and conversion rate after the deployment further confirms the impact on key business metrics. We believe that the flexibility of our framework makes it a promising solution toward enriching the user experience across the e-commerce landscape. |
| title | DashCLIP: Leveraging multimodal models for generating semantic embeddings for DoorDash |
| topic | Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2504.07110 |