Video Question Answering on STAR (val)
63.8Mean ScoreVideoChat2
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VideoChat2Params=7B, Zero-shot=true, Pre-trained on video=true2024.12 | 63.8 | 62.4 | 67.2 | 57.5 | 53.9 | |
| MISTBackbone=CLIP2022.12 | 51.13 | 55.59 | 54.23 | 54.24 | 44.48 | |
| MIST2024.01 | 51.13 | 55.59 | 54.23 | 54.24 | 44.48 | |
| VidCtxParams=7B, Zero-shot=true2024.12 | 51.1 | 53.9 | 54.3 | 51.4 | 44.7 | |
| MISTBackbone=AIO2022.12 | 49.69 | 53 | 52.37 | 49.52 | 43.87 | |
| Temp[ATP]2022.12 | 48.37 | 50.63 | 52.87 | 49.36 | 40.61 | |
| Temp[ATP]2024.01 | 48.37 | 50.63 | 52.87 | 49.36 | 40.61 | |
| AIO2022.12 | 47.54 | 47.53 | 50.81 | 47.75 | 44.08 | |
| AIO2024.01 | 47.54 | 47.53 | 50.81 | 47.75 | 44.08 | |
| Q-ViDParams=12B, Zero-shot=true2024.12 | 45.7 | 48.2 | 47.2 | 43.9 | 43.4 | |
| TraveLERzero-shot=true2024.04 | 44.9 | — | — | — | — | |
| SeViLAzero-shot=true, fine-tuned components=true2024.04 | 44.6 | — | — | — | — | |
| SeViLAParams=4B, Zero-shot=true, Pre-trained on video=true2024.12 | 44.6 | 48.3 | 45 | 44.4 | 40.8 | |
| FlamingoParameters=9B2022.12 | 43.4 | — | — | — | — | |
| BLIP-2 concatzero-shot=true, aggregation=concat2024.04 | 42.2 | — | — | — | — | |
| Flamingo-9Bzero-shot=true2024.04 | 41.8 | — | — | — | — | |
| InternVideozero-shot=true2024.04 | 41.6 | — | — | — | — | |
| InternVideoParams=<1B, Zero-shot=true, Pre-trained on video=true2024.12 | 41.6 | 43.8 | 43.2 | 42.3 | 37.4 | |
| RESERVE-B2022.12 | 40.5 | 44.8 | 42.4 | 38.8 | 36.2 | |
| BLIP-2 votingzero-shot=true, aggregation=voting2024.04 | 40.3 | — | — | — | — | |
| CLIP2022.12 | 38 | 39.8 | 40.5 | 35.5 | 36 | |
| ClipBERT2022.12 | 36.7 | 39.81 | 43.59 | 32.34 | 31.42 |