Trajectory Evaluation via Rollout-Free World Model for End-to-end Autonomous Driving
This work proposes RFWorld, a Rollout-Free World model for end-to-end trajectory evaluation, which constructs a shared scene memory and shapes it into a future-readable representation via lightweight training-time temporal readout with semantic supervision, keeping training overhead low.
Jun-Zhe Wang, Jin-Qing Zhang, Qing-Jie Liu
· 0 citations