'Does it come in black?' CLIP-like models are zero-shot recommenders

Autor:	Chia, Patrick John, Tagliabue, Jacopo, Bianchi, Federico, Greco, Ciro, Goncalves, Diogo
Rok vydání:	2022
Předmět:	Computer Science - Information Retrieval Computer Science - Artificial Intelligence
Druh dokumentu:	Working Paper
Popis:	Product discovery is a crucial component for online shopping. However, item-to-item recommendations today do not allow users to explore changes along selected dimensions: given a query item, can a model suggest something similar but in a different color? We consider item recommendations of the comparative nature (e.g. "something darker") and show how CLIP-based models can support this use case in a zero-shot manner. Leveraging a large model built for fashion, we introduce GradREC and its industry potential, and offer a first rounded assessment of its strength and weaknesses. Comment: Accepted at ACL 2022 (ECNLP)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2204.02473 Zobrazit plný text záznamu View this record from Arxiv