List your company

Speech-Visual Question Answering on A-OKVQA Speech-converted

0.2001AccuracyUni-MoE
0.001460.053030.10460.15617May 18, 2024

Evaluation results

MethodLinks
Uni-MoETraining Task=MoE-Task1-short-speech, Training Epochs=3 epoch2024.050.2001
Uni-MoETraining Task=MoE-Task1-short-speech, Training Epochs=2 epoch2024.050.1931
Single-Modality-Expert-Task3Evaluation Protocol=zero-shot2024.050.1926
Single-Modality-Expert-Task1Evaluation Protocol=zero-shot2024.050.1838
Uni-MoETraining Task=MoE-Task1, Training Epochs=2 epoch2024.050.148
Uni-MoETraining Task=MoE-Task1, Training Epochs=3 epoch2024.050.1358
Single-Modality-Expert-Task6Evaluation Protocol=zero-shot2024.050.114
Single-Modality-Expert-Task4Evaluation Protocol=zero-shot2024.050.0914
Uni-MoETraining Task=MoE-Task2, Training Epochs=2 epoch2024.050.0776
Single-Modality-Expert-Task5Evaluation Protocol=zero-shot2024.050.0456
Macaw-LLMEvaluation Protocol=zero-shot2024.050.0108
X-InstructBLIPEvaluation Protocol=zero-shot2024.050.0091