Speech Emotion Recognition on M3ED
32.1Macro F1B-GRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| B-GRPOTraining Protocol=RL-based unsupervised (after 100 epoch supervised pre-training)2026.02 | 32.1 | |
| Full labeledTraining Protocol=Supervised (full dataset)2026.02 | 31.5 | |
| DINOTraining Protocol=Unsupervised learning2026.02 | 30.8 | |
| Same epochsTraining Protocol=Supervised (200 epochs)2026.02 | 29.7 | |
| BaselineTraining Protocol=Supervised (half labeled, 100 epochs)2026.02 | 28.8 |