Hugging Face's logo

JorgeAV
/
MR-JEPA

multimodal
reasoning
jepa
world-model
vision-language
Model card Files
xet
 Community
MR-JEPA / mr_jepa /models /target_encoder.py

Commit History

fix: target_encoder.py — respect config.jepa_loss_fn (smooth_l1/mse/cosine) instead of hardcoded MSE
ac52da3
verified

JorgeAV commited on

Initial MR-JEPA codebase: architecture, training, evaluation, and tests
dba2c56
verified

JorgeAV commited on