Video Question Answering on DramaQA
84.1Total AccuracyLLaMA-VQA
Evaluation results
| Method | Links | |
|---|---|---|
| LLaMA-VQALanguage Model=LLaMA, Number of Trainable Parameters=4.5M2023.10 | 84.1 | |
| FrozenBiLMLanguage Model=DeBERTa, Number of Trainable Parameters=30M2023.10 | 82 | |
| MERLOTLanguage Model=ROBERTa, Number of Trainable Parameters=223M2023.10 | 81.4 | |
| SPCRLLanguage Model=BERT2023.10 | 81 | |
| VidLLanguage Model=BERT, Number of Trainable Parameters=25M2023.10 | 79 |