Audio Visual Grounding on SoundNet
88.4cIoUMeerKat
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MeerKatModel Category=Task Specific MLLMs2026.04 | 88.4 | 67.9 | |
| BuboGPTModel Category=Omnimodal MLLMs2026.04 | 81.2 | 62.3 | |
| Qwen3-OmniModel Category=Omnimodal MLLMs2026.04 | 39.2 | 41.7 | |
| Omni-o32026.04 | 27.5 | 28.8 | |
| Qwen2.5-OmniModel Category=Omnimodal MLLMs2026.04 | 6.6 | 9.3 |