VL-JEPA: Joint Embedding Predictive Architecture for Vision-Language
hbarka Wednesday, December 31, 2025
2
0
arxiv.org
Story