Researcher Says JEPA-Titled Models Mostly Underperform, V-JEPA 2.1 Excepted

Gabriele Berton sparked a small AI-community debate, claiming models with JEPA in the title usually underperform—LeJEPA trails DINO—except possibly V-JEPA 2.1, with some suspecting the naming aims to attract LeCun's reposts.

2026-11-07 ~ 2026-11-07 · 2 related posts

• Researcher: JEPA-titled models are usually bad in practice, V-JEPA 2.1 the only exception — serrjoa · 2026-11-07
• Researcher claims JEPA-titled models underperform DINO, mocks LeCun shoutout chasing — gabriberton · 2026-11-07