AI RESEARCH
Text-Conditional JEPA for Learning Semantically Rich Visual Representations
arXiv CS.LG
•
ArXi:2605.03245v1 Announce Type: new Image-based Joint-Embedding Predictive Architecture (I-JEPA) offers a promising approach to visual self-supervised learning through masked feature prediction. However with the inherent visual uncertainty at masked positions, feature prediction remains challenging and may fail to learn semantic representations. In this work, we propose Text-Conditional JEPA (TC-JEPA) that uses image captions to reduce the prediction uncertainty.