Story Generation on Claude-3 Story Evaluation Scripts via Dramatron (test)
78.97FascinationMoPS-Dramatron
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MoPS-DramatronPremise source=MoPS, Downstream generator=Dramatron, Evaluator model=Claude-3-opus2024.06 | 78.97 | 82.5 | 86.17 | 82.54 | |
| CPX-DramatronPremise source=CPX, Downstream generator=Dramatron, Evaluator model=Claude-3-opus2024.06 | 78.33 | 81.83 | 86.17 | 82.11 | |
| STM-DramatronPremise source=STM, Downstream generator=Dramatron, Evaluator model=Claude-3-opus2024.06 | 73.5 | 76.17 | 82.83 | 77.5 | |
| WP-DramatronPremise source=WP, Downstream generator=Dramatron, Evaluator model=Claude-3-opus2024.06 | 72.76 | 73.33 | 82.17 | 76.09 |