{
  "title": "V-JEPA predicts held-out features",
  "caption": "V-JEPA learns to predict held-out target embeddings from visible video context. The target encoder uses an exponential moving average (EMA), and its features are stop-gradient targets. Author-redrawn mechanism from Bardes et al., without token grids and implementation shapes. Source: Bardes et al. Playback time is illustrative and is not measured runtime.",
  "alt": "V-JEPA training flow: context predicts held-out features, and an EMA encoder gives stop-gradient targets.",
  "sources": [
    "https://arxiv.org/abs/2404.08471"
  ],
  "limitations": [
    "This playback is an example. Playback duration is not measured runtime.",
    "The diagram does not include detailed optimizer and reverse-gradient paths for context/predictor parameters.",
    "The target encoder receives EMA weights, not the feature-loss gradient."
  ]
}