{"ID":2888217,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.05661","arxiv_id":"2508.05661","title":"Zero-Shot Retrieval for Scalable Visual Search in a Two-Sided Marketplace","abstract":"Visual search offers an intuitive way for customers to explore diverse product catalogs, particularly in consumer-to-consumer (C2C) marketplaces where listings are often unstructured and visually driven. This paper presents a scalable visual search system deployed in Mercari's C2C marketplace, where end-users act as buyers and sellers. We evaluate recent vision-language models for zero-shot image retrieval and compare their performance with an existing fine-tuned baseline. The system integrates real-time inference and background indexing workflows, supported by a unified embedding pipeline optimized through dimensionality reduction. Offline evaluation using user interaction logs shows that the multilingual SigLIP model outperforms other models across multiple retrieval metrics, achieving a 13.3% increase in nDCG@5 over the baseline. A one-week online A/B test in production further confirms real-world impact, with the treatment group showing substantial gains in engagement and conversion, up to a 40.9% increase in transaction rate via image search. Our findings highlight that recent zero-shot models can serve as a strong and practical baseline for production use, which enables teams to deploy effective visual search systems with minimal overhead, while retaining the flexibility to fine-tune based on future data or domain-specific needs.","short_abstract":"Visual search offers an intuitive way for customers to explore diverse product catalogs, particularly in consumer-to-consumer (C2C) marketplaces where listings are often unstructured and visually driven. This paper presents a scalable visual search system deployed in Mercari's C2C marketplace, where end-users act as bu...","url_abs":"https://arxiv.org/abs/2508.05661","url_pdf":"https://arxiv.org/pdf/2508.05661v1","authors":"[\"Andre Rusli\",\"Shoma Ishimoto\",\"Sho Akiyama\",\"Aman Kumar Singh\"]","published":"2025-07-31T05:13:20Z","proceeding":"cs.IR","tasks":"[\"cs.IR\",\"cs.AI\"]","methods":"[\"Language Model\"]","has_code":false}