Video Captioning on MiraData-9k 1.0 (test)
81.6HalFscore (Object - Hal)HALVA
Evaluation results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HALVAbase_model=LLaVA-Video2025.12 | 81.6 | 84.5 | 50 | 22.9 | 26.9 | 18.5 | 67 | 74.4 | 51.5 | 22.8 | 40.2 | 24.2 | |
| LLaVA-Videobase_model=LLaVA-Video2025.12 | 79.4 | 76.5 | 52.1 | 33 | 29.5 | 27.5 | 49 | 58.4 | 54.6 | 32.9 | 52.7 | 36.7 | |
| HALVAbase_model=LLaVA-Video2025.12 | 78.9 | 75 | 50.5 | 29.4 | 29.8 | 27 | 48.7 | 58 | 53.5 | 28.5 | 52.4 | 34 | |
| LLaVA-Videobase_model=LLaVA-Video2025.12 | 78.7 | 85.4 | 42.3 | 19.4 | 38.9 | 22.3 | 65.1 | 73 | 44 | 19 | 38.9 | 22.3 | |
| DeCobase_model=LLaVA-Video2025.12 | 77.1 | 86.2 | 40 | 16.2 | 29 | 14.9 | 65.2 | 73.4 | 41.3 | 15.9 | 37.8 | 19.9 | |
| DeCobase_model=LLaVA-Video2025.12 | 74.2 | 73.4 | 46.6 | 27.3 | 33.2 | 26.9 | 48.4 | 56.6 | 48.1 | 27.6 | 49.8 | 33.8 | |
| SFTbase_model=LLaVA-Video2025.12 | 72 | 67.6 | 49.7 | 23.3 | 35.8 | 27.1 | 44.9 | 54.6 | 54.7 | 22.9 | 54.9 | 30.4 | |
| HACLbase_model=LLaVA-Video2025.12 | 70 | 65.2 | 53.7 | 29.9 | 38.4 | 32.2 | 42.7 | 51.6 | 60.8 | 28.6 | 59 | 35.9 | |
| SANTAbase_model=LLaVA-Video2025.12 | 69.6 | 64.5 | 57.2 | 32.6 | 39.7 | 34 | 40.9 | 50.3 | 64.1 | 31.5 | 61.5 | 38.6 | |
| SFTbase_model=LLaVA-Video2025.12 | 63.3 | 70.1 | 43 | 18.2 | 39.6 | 22.6 | 63.4 | 70.7 | 38.2 | 14.3 | 37.4 | 19.2 | |
| SANTAbase_model=LLaVA-Video2025.12 | 62.2 | 68.3 | 47.2 | 22.8 | 42.3 | 25.8 | 60.2 | 68.4 | 48.6 | 22.3 | 43.8 | 26.1 | |
| HACLbase_model=LLaVA-Video2025.12 | 60.3 | 65.7 | 43.5 | 18.8 | 41.5 | 24.3 | 60.4 | 68.5 | 44.7 | 18.4 | 42 | 23.2 |