DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving

Understand

We propose DynVLA, a driving VLA model that introduces a new CoT paradigm termed Dynamics CoT.

• DynVLA forecasts compact world dynamics before action generation, enabling more informed and physically grounded decision-making.
• To obtain compact dynamics representations, DynVLA introduces a Dynamics Tokenizer that compresses future evolution into a small set of dynamics tokens.
• Considering the rich environment dynamics in interaction-intensive driving scenarios, DynVLA decouples ego-centric and environment-centric dynamics, yielding more accurate world dynamics modeling.