Speech-Visual Question Answering on A-OKVQA Speech-converted
0.2001AccuracyUni-MoE
Evaluation Results
| Method | Links | |
|---|---|---|
| Uni-MoETraining Task=MoE-Task1-short-speech, Training Epochs=3 epoch2024.05 | 0.2001 | |
| Uni-MoETraining Task=MoE-Task1-short-speech, Training Epochs=2 epoch2024.05 | 0.1931 | |
| Single-Modality-Expert-Task3Evaluation Protocol=zero-shot2024.05 | 0.1926 | |
| Single-Modality-Expert-Task1Evaluation Protocol=zero-shot2024.05 | 0.1838 | |
| Uni-MoETraining Task=MoE-Task1, Training Epochs=2 epoch2024.05 | 0.148 | |
| Uni-MoETraining Task=MoE-Task1, Training Epochs=3 epoch2024.05 | 0.1358 | |
| Single-Modality-Expert-Task6Evaluation Protocol=zero-shot2024.05 | 0.114 | |
| Single-Modality-Expert-Task4Evaluation Protocol=zero-shot2024.05 | 0.0914 | |
| Uni-MoETraining Task=MoE-Task2, Training Epochs=2 epoch2024.05 | 0.0776 | |
| Single-Modality-Expert-Task5Evaluation Protocol=zero-shot2024.05 | 0.0456 | |
| Macaw-LLMEvaluation Protocol=zero-shot2024.05 | 0.0108 | |
| X-InstructBLIPEvaluation Protocol=zero-shot2024.05 | 0.0091 |