Here is the training data for paper:"Unified Vision-Language Representation Modeling for E-Commerce Same-Style Products Retrieval". (Accepted in WWW2023 Industry track).

All the data are the same with those for the experiments of esspr. 

The data format is formulated as: 'cate_lv1_name,cate_name,trigger_id,tirgger_title,recall_id,recall_title,trigger_img,recall_img,label'

All the image and title for each items are online grapped, with only id_value is virtual, in order to protect privacy.

The images can be directly downloaded without any limitation.

If you use this dataset, please cite this articles. Thanks.