Autonomous Driving Question Answering on LingoQA (val)
61Lingo-JLEO
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LEON (number of frames used during training)=22025.01 | 61 | 14.91 | 35.44 | 69.72 | |
| LingoQAN (number of frames used during training)=52025.01 | 60.8 | 15 | 18.56 | 65.62 | |
| LingoQAN (number of frames used during training)=32025.01 | 59.8 | 14.61 | 18.44 | 62.61 | |
| InternVLN (number of frames used during training)=52025.01 | 58 | 13.53 | 34.27 | 67.17 | |
| BLIP 2N (number of frames used during training)=12025.01 | 52.2 | 13 | 17.4 | 60.1 | |
| LLaVA-1.5N (number of frames used during training)=52025.01 | 51 | 10.62 | 29.44 | 48.18 |