Physical Diagram Generation on PhyDrawGen E&M (Evaluation Set)
69Preference RatePhyDrawGen
Evaluation Results
| Method | Links | |
|---|---|---|
| PhyDrawGenEvaluation Protocol=Human-judge forced-choice comparison2026.05 | 69 | |
| GPT-5Evaluation Protocol=Human-judge forced-choice comparison2026.05 | 15.3 | |
| Gemini 3 ProEvaluation Protocol=Human-judge forced-choice comparison2026.05 | 11.9 | |
| Gemini 2.5 FlashEvaluation Protocol=Human-judge forced-choice comparison2026.05 | 3.8 |