Role-playing on Role-playing evaluation (Minor characters)
21.21K-On! ROUGE-LCodified Profile
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Codified ProfileRP Model=gpt-4.1, Discriminator Model=gpt-4.1-mini2026.02 | 21.21 | — | 18.09 | 21.4 | 21.66 | 19.17 | 20.31 | |
| Codified FSMRP Model=gpt-4.1, Discriminator Model=gpt-4.1-mini2026.02 | 20.93 | — | 18.74 | 22.88 | 21.49 | 20.02 | 20.81 | |
| Plot SummaryRP Model=gpt-4.1, Discriminator Model=gpt-4.1-mini2026.02 | 20.19 | — | 18.38 | 21.16 | 21.67 | 18.93 | 20.07 | |
| PromptTransRP Model=gpt-4.1, Discriminator Model=gpt-4.1-mini2026.02 | 20.18 | — | 18.13 | 20.78 | 21.27 | 19.06 | 19.88 | |
| Character UpdatingRP Model=gpt-4.1, Discriminator Model=gpt-4.1-mini2026.02 | 19.51 | — | 18.39 | 21.61 | 20.81 | 18.87 | 19.84 |