Plot Beat Prediction on Plot-Beat Serialized-Fiction (held-out)
93F1 Score (Plot-Beat)GPT-5
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5Horizon (h)=102026.06 | 93 | |
| Claude Opus 4.5Horizon (h)=102026.06 | 92 | |
| o3-ProHorizon (h)=102026.06 | 92 | |
| Gemini-2.5-ProHorizon (h)=102026.06 | 91 | |
| N-VSSMHorizon (h)=102026.06 | 89 | |
| GPT-5Horizon (h)=202026.06 | 88 | |
| N-VSSMHorizon (h)=202026.06 | 88 | |
| Claude Opus 4.5Horizon (h)=202026.06 | 87 | |
| o3-ProHorizon (h)=202026.06 | 87 | |
| N-VSSMHorizon (h)=502026.06 | 86 | |
| Gemini-2.5-ProHorizon (h)=202026.06 | 85 | |
| N-VSSMHorizon (h)=1002026.06 | 85 | |
| N-VSSMHorizon (h)=2002026.06 | 84 | |
| GPT-5Horizon (h)=502026.06 | 81 | |
| Claude Opus 4.5Horizon (h)=502026.06 | 80 | |
| Gemini-2.5-ProHorizon (h)=502026.06 | 79 | |
| o3-ProHorizon (h)=502026.06 | 79 | |
| GPT-5Horizon (h)=1002026.06 | 76 | |
| Claude Opus 4.5Horizon (h)=1002026.06 | 74 | |
| GPT-5Horizon (h)=2002026.06 | 73 | |
| o3-ProHorizon (h)=1002026.06 | 73 | |
| Gemini-2.5-ProHorizon (h)=1002026.06 | 72 | |
| Claude Opus 4.5Horizon (h)=2002026.06 | 71 | |
| Gemini-2.5-ProHorizon (h)=2002026.06 | 71 | |
| o3-ProHorizon (h)=2002026.06 | 71 |