Visual Question Answering on nuScenes-QA
61.3Overall ScoreHERMES++
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| HERMES++Modality=Camera2026.04 | 61.3 | — | — | — | — | — | — | — | |
| CenterPoint+MCANModality=LiDAR2026.04 | 59.5 | — | — | — | — | — | — | — | |
| Omni-QModality=Camera2026.04 | 59.2 | — | — | — | — | — | — | — | |
| OpenDriveVLA-0.5BModel size=0.5B2026.05 | 58.4 | 62.3 | 56.5 | 83.9 | 22 | 50.2 | 57 | 68.4 | |
| OpenDriveVLA-7BModality=Camera2026.04 | 58.2 | — | — | — | — | — | — | — | |
| CenterPoint+BUTDModality=LiDAR2026.04 | 58.1 | — | — | — | — | — | — | — | |
| BEVDet+MCANModality=Camera2026.04 | 57.9 | — | — | — | — | — | — | — | |
| BEVDet+BUTDModality=Camera2026.04 | 57 | — | — | — | — | — | — | — | |
| BEVDet+BUTD2026.05 | 57 | — | — | 83.7 | 20.9 | 48.8 | 52 | 67.7 | |
| Ours-0.5BModel size=0.5B2026.05 | 56.4 | 59.9 | 54.7 | 83.7 | 19.5 | 47.1 | 52.8 | 67.3 | |
| MVLAD-AD2026.02 | 55.7 | 58.5 | 54.3 | — | — | — | — | — | |
| ALN-P32026.02 | 52.9 | 57.1 | 50.9 | — | — | — | — | — | |
| Hint-AD2026.02 | 50.5 | 55.4 | 48 | — | — | — | — | — | |
| TOD3Cap2026.02 | 49 | 53 | 45.1 | — | — | — | — | — | |
| LiDAR-LLMModality=LiDAR2026.04 | 48.6 | — | — | — | — | — | — | — | |
| LiDAR-LLM2026.05 | 48.6 | — | — | 74.5 | 15 | 37.8 | 45.9 | 57.8 | |
| LLaVAModality=Camera2026.04 | 47.4 | — | — | — | — | — | — | — | |
| GPT-4o2026.02 | 37.1 | 42 | 34.7 | — | — | — | — | — | |
| Gemini-1.52026.02 | 35.4 | 40.5 | 32.9 | — | — | — | — | — | |
| LLaVA1.52026.05 | 26.2 | 25.7 | 41.5 | 45.8 | 7.7 | 7.8 | 9 | 52.1 | |
| LLaMA-AdapV22026.05 | 9.6 | 15.1 | 4.8 | 19.3 | 2.7 | 7.6 | 10.8 | 1.6 |