MRSE: An Efficient Multi-modality Retrieval System for Large Scale E-commerce
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912002945843200 |
|---|---|
| author | Jiang, Hao Zhang, Haoxiang Hou, Qingshan Chen, Chaofeng Lin, Weisi Zhang, Jingchang Wang, Annan |
| author_facet | Jiang, Hao Zhang, Haoxiang Hou, Qingshan Chen, Chaofeng Lin, Weisi Zhang, Jingchang Wang, Annan |
| contents | Providing high-quality item recall for text queries is crucial in large-scale e-commerce search systems. Current Embedding-based Retrieval Systems (ERS) embed queries and items into a shared low-dimensional space, but uni-modality ERS rely too heavily on textual features, making them unreliable in complex contexts. While multi-modality ERS incorporate various data sources, they often overlook individual preferences for different modalities, leading to suboptimal results. To address these issues, we propose MRSE, a Multi-modality Retrieval System that integrates text, item images, and user preferences through lightweight mixture-of-expert (LMoE) modules to better align features across and within modalities. MRSE also builds user profiles at a multi-modality level and introduces a novel hybrid loss function that enhances consistency and robustness using hard negative sampling. Experiments on a large-scale dataset from Shopee and online A/B testing show that MRSE achieves an 18.9% improvement in offline relevance and a 3.7% gain in online core metrics compared to Shopee's state-of-the-art uni-modality system. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_14968 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MRSE: An Efficient Multi-modality Retrieval System for Large Scale E-commerce Jiang, Hao Zhang, Haoxiang Hou, Qingshan Chen, Chaofeng Lin, Weisi Zhang, Jingchang Wang, Annan Information Retrieval Computation and Language Providing high-quality item recall for text queries is crucial in large-scale e-commerce search systems. Current Embedding-based Retrieval Systems (ERS) embed queries and items into a shared low-dimensional space, but uni-modality ERS rely too heavily on textual features, making them unreliable in complex contexts. While multi-modality ERS incorporate various data sources, they often overlook individual preferences for different modalities, leading to suboptimal results. To address these issues, we propose MRSE, a Multi-modality Retrieval System that integrates text, item images, and user preferences through lightweight mixture-of-expert (LMoE) modules to better align features across and within modalities. MRSE also builds user profiles at a multi-modality level and introduces a novel hybrid loss function that enhances consistency and robustness using hard negative sampling. Experiments on a large-scale dataset from Shopee and online A/B testing show that MRSE achieves an 18.9% improvement in offline relevance and a 3.7% gain in online core metrics compared to Shopee's state-of-the-art uni-modality system. |
| title | MRSE: An Efficient Multi-modality Retrieval System for Large Scale E-commerce |
| topic | Information Retrieval Computation and Language |
| url | https://arxiv.org/abs/2408.14968 |