API
Download
🥇 Rank 1
Winner video
STLA-MINES
Caio Azevedo, Stefano Sabatini, Sascha Hornauer, Fabien Moutarde
Stellantis and Mines Paris
MMS
5.15
Semantic coherence
N/A
End-to-end driving on 1,000 rare scenarios. Ranked by Multi-Maneuver Score (MMS, 0–10) — a metric significantly more correlated with closed-loop DrivingScore than standard L2 error. Best submission per method shown.
Selected challenge winners, paired with LongTail driving clips.
Winner video
Caio Azevedo, Stefano Sabatini, Sascha Hornauer, Fabien Moutarde
Stellantis and Mines Paris
MMS
5.15
Semantic coherence
N/A
No video available
Runner-up video
Sanath Tiptur Sadashivaiah, Taehyoung, Abhishek
TH Aschaffenburg, Fraunhofer IVI, TH Ingolstadt
MMS
4.31
Semantic coherence
0.84
Third-place video
Yijie Wang, Kashyap Chitta
University of Toronto, KE:SAI
MMS
4.31
Semantic coherence
0.39
Live · fetched from HuggingFace · best submission per method
Could not load leaderboard. View on HuggingFace ↗
MMS
↑ higher is better
Multi-Maneuver Score (0–10). Composite score covering trajectory accuracy and semantic compliance across scenario types. Defined in the KITScenes LongTail paper. arXiv:2603.23607 ↗
Semantic Coherence
↑ higher is better
Semantic Coherence measures whether the driving actions described in a model's reasoning trace match its planned trajectory. It is computed via an embedding-based Roccihio classification. Defined in the KITScenes LongTail paper. arXiv:2603.23607 ↗
