RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning Paper • 2502.13144 • Published Feb 18, 2025 • 38
