CausalDreamer: Learning Predictive World Models with Latent Disentanglement
CausalDreamer re-encodes a frozen Dreamer 4/MMBench2 video tokenizer latent into four groups factorized by controllability and reward relevance, routes actions only to the controllable groups, and predicts reward from only the reward-relevant groups. The pretrained dynamics model is fine-tuned to predict this factored representation. On MMBench2, it improves planning scores on clean tasks and their manipulated variants.
