Understanding on EventDrive
65.46AccuracyEventDrive-VLM
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| EventDrive-VLMInput Modality=Event + Frame, Evaluation Protocol=Fine-tuned2026.06 | 65.46 | 72.86 | 72.56 | |
| LLaVA-OneVision-1.5-8BInput Modality=Frame, Evaluation Protocol=Zero-shot2026.06 | 61.62 | 2.91 | 12.5 | |
| InternVL3-8BInput Modality=Frame, Evaluation Protocol=Zero-shot2026.06 | 60.6 | 0.24 | 2.31 | |
| InternVL2.5-8BInput Modality=Frame, Evaluation Protocol=Zero-shot2026.06 | 60.02 | 0.29 | 2.5 | |
| Qwen2.5-VL-7B-Instruct*Input Modality=Frame, Evaluation Protocol=Fine-tuned2026.06 | 58.44 | 69.94 | 67.19 | |
| EventDrive-VLMInput Modality=Event, Evaluation Protocol=Fine-tuned2026.06 | 54.21 | 43.43 | 47.82 | |
| Qwen2.5-VL-7B-InstructInput Modality=Frame, Evaluation Protocol=Zero-shot2026.06 | 49.98 | 63.35 | 60.11 | |
| LLaVA-v1.6-Mistral-7B-hfInput Modality=Frame, Evaluation Protocol=Zero-shot2026.06 | 40.37 | 15.02 | 29 | |
| EventGPT-7BInput Modality=Event, Evaluation Protocol=Zero-shot2026.06 | 38.78 | 5.49 | 8.24 |