Emotional Reasoning on Emotion6 Out-of-domain
60.44AccuracyEMO-R3
Evaluation Results
| Method | Links | |
|---|---|---|
| EMO-R3Backbone=Qwen2.5-VL-3B-Instruct, Roll-out=4, Post-training=true2026.02 | 60.44 | |
| GRPOBackbone=Qwen2.5-VL-3B-Instruct, Roll-out=4, Post-training=true2026.02 | 60.1 | |
| EMO-R3Backbone=Qwen2.5-VL-3B-Instruct, Roll-out=8, Post-training=true2026.02 | 59.26 | |
| GRPOBackbone=Qwen2.5-VL-3B-Instruct, Roll-out=8, Post-training=true2026.02 | 57.91 | |
| DAPOBackbone=Qwen2.5-VL-3B-Instruct, Roll-out=4, Post-training=true2026.02 | 56.9 | |
| DAPOBackbone=Qwen2.5-VL-3B-Instruct, Roll-out=8, Post-training=true2026.02 | 55.72 | |
| SEPM*Backbone=LLaVA1.5-7B, Roll-out=-, Post-training=false2026.02 | 54.21 | |
| Vanilla*Backbone=Qwen2.5-VL-3B-Instruct, Roll-out=-, Post-training=false2026.02 | 50 | |
| Vanilla*Backbone=LLaVA1.5-7B, Roll-out=-, Post-training=false2026.02 | 48.32 | |
| SFTBackbone=Qwen2.5-VL-3B-Instruct, Roll-out=-, Post-training=true2026.02 | 34.51 |