Video Question Answering on MSVD-QA LLaVA-Hound out-of-domain (test)
67.2AccuracyMM1.5-Video
Evaluation Results
| Method | Links | |
|---|---|---|
| MM1.5-VideoModel Scale=7B, Protocol=SFT2024.09 | 67.2 | |
| MM1.5-VideoModel Scale=3B, Protocol=SFT2024.09 | 65.2 | |
| MM1.5-VideoModel Scale=1B, Protocol=SFT2024.09 | 65 | |
| LLAVA-HOUND-SFTModel Scale=7B2024.09 | 62.2 | |
| MM1.5-VideoModel Scale=7B, Protocol=Training-free2024.09 | 52.9 | |
| MM1.5-VideoModel Scale=3B, Protocol=Training-free2024.09 | 51.1 | |
| MM1.5-VideoModel Scale=1B, Protocol=Training-free2024.09 | 49.8 | |
| Video-LLaVAModel Scale=7B2024.09 | 39.5 | |
| Chat-UniViModel Scale=7B2024.09 | 35.6 | |
| Video-ChatGPTModel Scale=7B2024.09 | 34.1 | |
| LLaMA-VIDModel Scale=7B2024.09 | 34.1 |