Single-Stream Multi-Level Alignment for Vision-Language Pretraining
收藏DataCite Commons2026-01-07 更新2026-05-05 收录
下载链接：
https://service.tib.eu/ldmservice/dataset/128526eb-69b1-4e8a-adef-6d0495d13767
下载链接
链接失效反馈官方服务：
资源简介：
Self-supervised vision-language pretraining from pure images and text with a contrastive loss is effective, but ignores fine-grained alignment due to a dual-stream architecture that aligns image and text representations only on a global level.
提供机构：
TIB
创建时间：
2024-12-16
