Question Answering on NExT-QA (Open-ended VideoQA)
Metric: Accuracy (higher is better)
Metric: Accuracy (higher is better)
| # | Model↕ | Accuracy▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | Flash-VStream | 61.6 | No | Flash-VStream: Memory-Based Real-Time Understand... | 2024-06-12 | Code |
| 2 | Vista-LLaMA | 60.7 | No | Vista-LLaMA: Reliable Video Narrator via Equal D... | 2023-12-12 | - |
| 3 | VideoChat | 56.6 | No | VideoChat: Chat-Centric Video Understanding | 2023-05-10 | Code |
| 4 | MovieChat+ | 54.8 | No | MovieChat+: Question-aware Sparse Memory for Lon... | 2024-04-26 | Code |
| 5 | Video-ChatGPT | 54.6 | No | Video-ChatGPT: Towards Detailed Video Understand... | 2023-06-08 | Code |
| 6 | MovieChat | 49.9 | No | MovieChat: From Dense Token to Sparse Memory for... | 2023-07-31 | Code |