Autonomous Driving on Intersection Collision critical scene
99.28Success RateSPHINX
Evaluation Results
| Method | Links | |
|---|---|---|
| SPHINXDriving backbone=RAP-ResNet2026.06 | 99.28 | |
| SPHINXDriving backbone=TCP2026.06 | 93.74 | |
| SPHINXDriving backbone=DAVE-22026.06 | 87.12 | |
| SPHINXDriving backbone=TransFuser2026.06 | 86.64 | |
| LLM-AttackerDriving backbone=TCP2026.06 | 70.5 | |
| LLM-AttackerDriving backbone=TransFuser2026.06 | 62.34 | |
| LLM-AttackerDriving backbone=RAP-ResNet2026.06 | 61.05 | |
| LLM-AttackerDriving backbone=DAVE-22026.06 | 53.95 | |
| ChatSceneDriving backbone=TCP2026.06 | 53.53 | |
| ChatSceneDriving backbone=TransFuser2026.06 | 47.22 | |
| ChatSceneDriving backbone=RAP-ResNet2026.06 | 47.01 | |
| ChatSceneDriving backbone=DAVE-22026.06 | 39.13 | |
| Base PerformanceDriving backbone=TCP2026.06 | 31.22 | |
| Base PerformanceDriving backbone=TransFuser2026.06 | 23.29 | |
| Base PerformanceDriving backbone=RAP-ResNet2026.06 | 22.84 | |
| Base PerformanceDriving backbone=DAVE-22026.06 | 16.62 |