Comprehensive Emotion Assessment on EEmo-Bench
0.6622Total Overall ScoreEEmo-Logic
Evaluation results
| Method | Links | |
|---|---|---|
| EEmo-Logic2026.02 | 0.6622 | |
| GPT-4o2026.02 | 0.6239 | |
| LLaVA-OneVision-72BScale=72B2026.02 | 0.6168 | |
| GPT-52026.02 | 0.6057 | |
| Claude-3.7-Sonnet2026.02 | 0.6014 | |
| Qwen2.5-VL-72BScale=72B2026.02 | 0.5913 | |
| Qwen2-VL-72BScale=72B2026.02 | 0.5869 | |
| Qwen-VL-Max2026.02 | 0.5811 | |
| Gemini-2.5-Pro2026.02 | 0.5714 | |
| Qwen2-VL-7BScale=7B2026.02 | 0.5605 | |
| Qwen2.5-VL-7BScale=7B2026.02 | 0.5571 | |
| Qwen3-VL-8BScale=8B2026.02 | 0.5534 | |
| LLaVA-OneVision-1.5-8BScale=8B2026.02 | 0.5524 | |
| mPLUG-Owl3-7BScale=7B2026.02 | 0.5511 | |
| LLaVA-OneVision-7BScale=7B2026.02 | 0.5455 | |
| InternVL3.5-8BScale=8B2026.02 | 0.5361 | |
| Emotion-Qwen2026.02 | 0.5333 | |
| EmoCaliber2026.02 | 0.5185 | |
| Janus-Pro-7BScale=7B2026.02 | 0.5163 | |
| Deepseek-VL-7B-chatScale=7B2026.02 | 0.5116 | |
| LLaVA-NEXT-8BScale=8B2026.02 | 0.4937 | |
| EmoVIT2026.02 | 0.3442 | |
| Random guess w/o open-ended2026.02 | 0.3365 | |
| R1-Omni-0.5BScale=0.5B2026.02 | 0.2553 | |
| AffectGPT2026.02 | 0.2241 |