Video Question Answering on MIMIC EchoQA (test)
50.8AccuracyEcho-CoPilot
Evaluation results
| Method | Links | |
|---|---|---|
| Echo-CoPilotBackbone=GPT-5.12025.12 | 50.8 | |
| Qwen2-VL-7B-biomed2025.12 | 49 | |
| o4-mini2025.12 | 43.9 | |
| Qwen2-VL-2B-biomed2025.12 | 42 | |
| GPT-4o2025.12 | 41.6 | |
| Phi-3.5-vision-instruct2025.12 | 41.1 | |
| InternVideo2.5-Chat-8B2025.12 | 40.3 | |
| Gemini-2.0-Flash2025.12 | 38.4 | |
| Qwen2-VL-7B-Instruct2025.12 | 37.9 | |
| Phi-4-multimodal-instruct2025.12 | 37.8 | |
| Qwen2-VL-72B-Instruct2025.12 | 37.5 | |
| Qwen2.5-VL-72B-Instruct2025.12 | 34.2 | |
| Qwen2.5-VL-7B-Instruct2025.12 | 34 | |
| MedGemma-4BPre-processing=EchoNet-based key-frame selector2025.12 | 33.4 | |
| Video-LLaVA2025.12 | 32 | |
| Video-ChatGPT2025.12 | 31.7 |