SOURCE-LINKED INTELLIGENCE

Frozen Flows Forget: Diagnosing and Restoring Lost Motion in a Latent-flow World Model

arXiv · AI, language, vision and robotics · article · Sep 23, 2026 · UTC

Latent world models that integrate a flow in a frozen self supervised latent space train stably and cheaply, yet silently lose the property manipulation depends on most: motion. The pretrained flow never moves the manipulated object; retraining it with latent-only losses only trades stillness for teleport-like motion. We trace the failure to the training signal, not the representation: anchor-sparse, latent-only supervision never says where along the horizon change belongs. Decode-augmented rollout training (DART) repairs this while keeping the representation frozen, retraining only the flow w

Read original source ↗ Open in workspace

recordType

paper

region

Global