Oops, SIGReg did it again! Large scale (CC12M->Datacomp-L) vision-language JEPA pretraining beats CLIP and SigLIP objectives! Thanks to SIGR…

Oops, SIGReg did it again! Large scale (CC12M->Datacomp-L) vision-language JEPA pretraining beats CLIP and SigLIP objectives! Thanks to SIGReg, our LeVLJEPA has no collapse, no EMA, no stop-gradient, no negatives, no problem! Checkpoints/demo are live: http://levljepa.github.io 🔥 We introduce LeVLJEPA: the first fully non-contrastive end-to-end vision-language pretraining method competitive with CLIP & SigLIP 💪🏼 👀 No negatives. No temperature. No momentum encoder. No teacher-student. TL;DR: LeVLJEPA learns image to text structure by prediction: each…

Source: Yann LeCun (X) | 2026-07-02