arXiv cs.AI
· Papers
TerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale
arXiv:2607.13028v2 Announce Type: replace-cross Abstract: Training robust autonomous driving agents requires a simulator fast enough for reinforcement learning at scale, realistic enough to ground behavior in real-world map structure, and diverse enough to cover the safety-critical long tail that logged data rarely con