Audio Question Answering on RACE Audio
64.21Accuracy (Sub-task 1)Uni-MoE w/ MoE-Task4 (4E)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Uni-MoE w/ MoE-Task4 (4E)Modality=I,T,S,V, Training Task Strategy=MoE-Task4, Number of Experts=42024.05 | 64.21 | 64.72 | |
| Uni-MoE w/ MoE-Task4 (4E) + Aux lossModality=I,T,S,V, Training Task Strategy=MoE-Task4, Number of Experts=4, Auxiliary Balancing Loss=true2024.05 | 63.86 | 64.24 | |
| Uni-MoE w/ MoE-Task4 (8E)Modality=I,T,S,V, Training Task Strategy=MoE-Task4, Number of Experts=82024.05 | 63.75 | 61.56 | |
| Uni-MoE w/ MoE-Task4 (8E) + Aux lossModality=I,T,S,V, Training Task Strategy=MoE-Task4, Number of Experts=8, Auxiliary Balancing Loss=true2024.05 | 62.33 | 64.18 | |
| Uni-MoE w/ MoE-Task2 (4E)Modality=I,T,S,V, Training Task Strategy=MoE-Task2, Number of Experts=42024.05 | 49.65 | 49.37 | |
| Uni-MoE w/ MoE-Task1 (4E)Modality=I,T,S, Training Task Strategy=MoE-Task1, Number of Experts=42024.05 | 47.08 | 47.08 | |
| Single-Modality-Expert-Task6Modality=I,T,S, Training Task Strategy=MoE-Task62024.05 | 32.59 | 29.02 | |
| Single-Modality-Expert-Task5Modality=I,T,S, Training Task Strategy=MoE-Task52024.05 | 30.78 | 24.9 | |
| X-InstructBLIPModality=Image, Text, Audio, Video2024.05 | 16.33 | 18.88 | |
| Macaw-LLMModality=Image, Text, Audio, Video2024.05 | 4.04 | 3 |