Human Correlation Analysis on Refined human judgment dataset human vs model-generated
0.995SO-SSO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SOHuman Judgment Criterion=Sensible/BT, Refinement Mode=Human vs Model-generated2023.05 | 0.995 | — | |
| SOHuman Judgment Criterion=Interesting/BT, Refinement Mode=Human vs Model-generated2023.05 | 0.81 | — | |
| SOHuman Judgment Criterion=Human-like/BT, Refinement Mode=Human vs Model-generated2023.05 | 0.357 | — | |
| MAUVEHuman Judgment Criterion=Human-like/BT, Refinement Mode=Human vs Model-generated2023.05 | — | 0.214 | |
| MAUVEHuman Judgment Criterion=Interesting/BT, Refinement Mode=Human vs Model-generated2023.05 | — | 0.667 | |
| MAUVEHuman Judgment Criterion=Sensible/BT, Refinement Mode=Human vs Model-generated2023.05 | — | 0.706 |