Audio Understanding / Audio Reasoning on MMAR
62.2AccuracyQwen2.5-Omni-7B + CoAT
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-Omni-7B + CoATBackbone=Qwen2.5-Omni-7B, Method=CoAT2026.06 | 62.2 | |
| MiMo-Audio + SICL-AT3few-shot=true2026.01 | 61.4 | |
| MiMo-Audio + SICL-AT2few-shot=true2026.01 | 61 | |
| Audio Flamingo 3 + CoATBackbone=Audio Flamingo 3, Method=CoAT2026.06 | 59.6 | |
| MiMo-Audio + Vanilla SICLfew-shot=true2026.01 | 58.2 | |
| MiMo-Audio + SICL-AT1few-shot=true2026.01 | 57.7 | |
| Qwen2.5-Omni-7BBackbone=Qwen2.5-Omni-7B, Method=Baseline2026.06 | 56.7 | |
| Audio Flamingo 3Backbone=Audio Flamingo 3, Method=Baseline2026.06 | 55.7 | |
| MiMo-Audiofew-shot=false2026.01 | 54.7 | |
| Qwen2.5-Omni + SICL-AT3few-shot=true2026.01 | 54.5 | |
| Qwen2.5-Omni + SICL-AT2few-shot=true2026.01 | 54.4 | |
| Qwen2.5-Omni + SICL-AT1few-shot=true2026.01 | 54.2 | |
| Qwen2.5-Omni + Vanilla SICLfew-shot=true2026.01 | 53.8 | |
| Qwen2-Audio + CoATBackbone=Qwen2-Audio, Method=CoAT2026.06 | 52.6 | |
| Qwen2.5-Omni (Fine-tuned on CV-en)few-shot=false2026.01 | 50.5 | |
| Qwen2.5-Omni (Fine-tuned on CV-en) + Vanilla SICLfew-shot=true2026.01 | 50.4 | |
| Qwen2.5-Omnifew-shot=false2026.01 | 49.2 | |
| Qwen2-AudioBackbone=Qwen2-Audio, Method=Baseline2026.06 | 47.1 | |
| Qwen2.5-Omni (Fine-tuned on RSR)few-shot=false2026.01 | 44.9 |