Model-Aware Generation of Safety-Critical Scenes for Robust End-to-End Driving via Distilled Planner Surrogates
End-to-end driving models perform well on standard benchmarks but collide frequently in safety-critical scenarios. We present a failure-mode analysis on NeuroNCAP which shows that 95.9% of DiffusionDrive's collisions are planning failures despite reliable detection, pinpointing the bottleneck in planning rather than pe...