Physical Plausibility Validation on PhysCodeBench (test)
4.6Gravity ScoreSMRF
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| SMRF2026.04 | 4.6 | 4.4 | 4.1 | 4.5 | 4.2 | 4.3 | |
| Claude-3.52026.04 | 3.8 | 3.5 | 3.2 | 3.7 | 3.6 | 3.6 |
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| SMRF2026.04 | 4.6 | 4.4 | 4.1 | 4.5 | 4.2 | 4.3 | |
| Claude-3.52026.04 | 3.8 | 3.5 | 3.2 | 3.7 | 3.6 | 3.6 |