Towards Better Search with Domain-Aware Text Embeddings for C2C Marketplaces
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915693325189120 |
|---|---|
| author | Rusli, Andre Cao, Miao Ishimoto, Shoma Akiyama, Sho Frenzel, Max |
| author_facet | Rusli, Andre Cao, Miao Ishimoto, Shoma Akiyama, Sho Frenzel, Max |
| contents | Consumer-to-consumer (C2C) marketplaces pose distinct retrieval challenges: short, ambiguous queries; noisy, user-generated listings; and strict production constraints. This paper reports our experiment to build a domain-aware Japanese text-embedding approach to improve the quality of search at Mercari, Japan's largest C2C marketplace. We experimented with fine-tuning on purchase-driven query-title pairs, using role-specific prefixes to model query-item asymmetry. To meet production constraints, we apply Matryoshka Representation Learning to obtain compact, truncation-robust embeddings. Offline evaluation on historical search logs shows consistent gains over a strong generic encoder, with particularly large improvements when replacing PCA compression with Matryoshka truncation. A manual assessment further highlights better handling of proper nouns, marketplace-specific semantics, and term-importance alignment. Additionally, an initial online A/B test demonstrates statistically significant improvements in revenue per user and search-flow efficiency, with transaction frequency maintained. Results show that domain-aware embeddings improve relevance and efficiency at scale and form a practical foundation for richer LLM-era search experiences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_21021 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Towards Better Search with Domain-Aware Text Embeddings for C2C Marketplaces Rusli, Andre Cao, Miao Ishimoto, Shoma Akiyama, Sho Frenzel, Max Information Retrieval Machine Learning Consumer-to-consumer (C2C) marketplaces pose distinct retrieval challenges: short, ambiguous queries; noisy, user-generated listings; and strict production constraints. This paper reports our experiment to build a domain-aware Japanese text-embedding approach to improve the quality of search at Mercari, Japan's largest C2C marketplace. We experimented with fine-tuning on purchase-driven query-title pairs, using role-specific prefixes to model query-item asymmetry. To meet production constraints, we apply Matryoshka Representation Learning to obtain compact, truncation-robust embeddings. Offline evaluation on historical search logs shows consistent gains over a strong generic encoder, with particularly large improvements when replacing PCA compression with Matryoshka truncation. A manual assessment further highlights better handling of proper nouns, marketplace-specific semantics, and term-importance alignment. Additionally, an initial online A/B test demonstrates statistically significant improvements in revenue per user and search-flow efficiency, with transaction frequency maintained. Results show that domain-aware embeddings improve relevance and efficiency at scale and form a practical foundation for richer LLM-era search experiences. |
| title | Towards Better Search with Domain-Aware Text Embeddings for C2C Marketplaces |
| topic | Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2512.21021 |