Menu

#Tipsv2

2 posts

Feed·
2 of 2 posts
TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment
📰
0

TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment

gdm-tipsv2.github.io·@HashtagPLUS·about 1 month ago
#jmrbmafj

TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment. A new family of image-text encoder models with strong dense patch-text alignment, evaluated across 9 tasks and 20 datasets.

15s
Read More