Audio Question Answering on MMAU v05.15.25 (test)
0.7374AccuracyAQA-TTRL
Evaluation Results
| Method | Links | |
|---|---|---|
| AQA-TTRLBackbone=Qwen2.5-Omni 7B, Protocol=Self-Adaptation2025.10 | 0.7374 | |
| DIMVBackbone=Qwen2.5-Omni 7B, Protocol=Direct Inference with Majority Voting2025.10 | 0.7201 | |
| SFTBackbone=Qwen2.5-Omni 7B, Protocol=Supervised Fine-Tuning2025.10 | 0.7174 | |
| AQA-TTRLBackbone=Qwen2.5-Omni 3B, Protocol=Self-Adaptation2025.10 | 0.7105 | |
| DIBackbone=Qwen2.5-Omni 7B, Protocol=Direct Inference2025.10 | 0.706 | |
| SFTBackbone=Qwen2.5-Omni 3B, Protocol=Supervised Fine-Tuning2025.10 | 0.6284 | |
| DIMVBackbone=Qwen2.5-Omni 3B, Protocol=Direct Inference with Majority Voting2025.10 | 0.6259 | |
| DIBackbone=Qwen2.5-Omni 3B, Protocol=Direct Inference2025.10 | 0.6155 |