Multimodal Emotion Reasoning on MME Emotion
57.3ER-LabMER-R1
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MER-R12026.06 | 57.3 | 49.6 | 38.7 | 37.2 | 42.2 | 72.1 | 33.9 | 62.9 | 38.4 | 64.6 | 51.5 | |
| OPPObackbone=Qwen2.5-Omni2026.06 | 54.6 | 45.5 | 42 | 40.7 | 39.6 | 66.2 | 34.8 | 60.6 | 31 | 68.1 | 49.5 | |
| VideoAuto-R12026.06 | 49.4 | 44.4 | 40.2 | 33.1 | 37.3 | 64.8 | 32.8 | 56.7 | 34.5 | 58.8 | 46.6 | |
| Baselinetraining=GRPO-style sampling, rewards=task and format rewards2026.06 | 47 | 40.8 | 35 | 35.6 | 37.7 | 59 | 32.4 | 58.7 | 27.9 | 62.7 | 45.3 | |
| Baseline2026.06 | 47 | 40.8 | 35 | 35.6 | 37.7 | 59 | 32.4 | 58.7 | 27.9 | 62.7 | 45.3 | |
| AffectGPT2026.06 | 35 | 32.8 | 32.6 | 44.2 | 31.3 | 50.7 | 30 | 47.9 | 19.7 | 57.4 | 38.6 | |
| AffectGPT2026.06 | 35 | 32.8 | 32.6 | 44.2 | 31.3 | 50.7 | 30 | 47.9 | 19.7 | 57.4 | 38.6 | |
| Emotion-LLaMA2026.06 | 31.7 | 19.7 | 31.7 | 36.4 | 29.5 | 49.3 | 27.3 | 40.7 | 21.1 | 42.4 | 31.8 | |
| Emotion-LLaMA2026.06 | 31.7 | 19.7 | 31.7 | 36.4 | 29.5 | 49.3 | 27.3 | 40.7 | 21.1 | 42.4 | 31.8 | |
| PandaGPT2026.06 | 28.5 | 23.2 | 25.2 | 41.4 | 28.9 | 26.3 | 29.2 | 37.7 | 21.6 | 38.4 | 30 | |
| PandaGPT2026.06 | 28.5 | 23.2 | 25.2 | 41.4 | 28.9 | 26.3 | 29.2 | 37.7 | 21.6 | 38.4 | 30 |