Role-Playing Evaluation (Social-Personality-Consistency) on SPC
66Win Rate (SPC)EBM
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| EBMOpponent=haruhi2026.05 | 66 | 17.8 | 16.2 | 48.2 | 78.8 | 1.22 | |
| EBMOpponent=crab2026.05 | 53.4 | 29.2 | 17.4 | 24.2 | 64.6 | 2.74 | |
| EBMOpponent=qwen2.5-vl-7b2026.05 | 51.6 | 27.4 | 21 | 24.2 | 65.3 | 1.17 | |
| EBMOpponent=qwen2.5-vl-7b-sft2026.05 | 48.4 | 32 | 19.6 | 16.4 | 60.2 | 5.06 | |
| EBMOpponent=intervl8b2026.05 | 47 | 32.4 | 20.6 | 14.6 | 59.2 | 2.91 | |
| EBMOpponent=RoleMRC2026.05 | 46 | 32.2 | 21.8 | 13.8 | 58.8 | 5.67 | |
| EBMOpponent=Internvl14b2026.05 | 44.8 | 36.2 | 19 | 8.6 | 55.3 | 0.0368 | |
| EBMOpponent=Qwen32b2026.05 | 35.2 | 40.6 | 24.2 | -5.4 | 46.4 | 0.1816 | |
| EBMOpponent=Internvl38b2026.05 | 33 | 48 | 19 | -15 | 40.7 | 2.27 |