Multimodal Physical Reasoning on WMW-TRACEBANK synthetic controlled split external-transfer pool
76Answer Accuracy
Claude Opus 4.7
Updated 5d ago
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
2026.05 | 76 | 81 | 68 | 91 | 18 | 3 | 5 | |
2026.05 | 72 | 77 | 61 | 88 | 24 | 4 | 6 | |
2026.05 | 63 | 68 | 49 | 83 | 31 | 4 | 7 | |
2026.05 | 60 | 65 | 47 | 82 | 33 | 3 | 6 | |
2026.05 | 58 | 63 | 44 | 80 | 34 | 3 | 5 | |
2026.05 | 52 | 55 | 38 | 78 | 35 | 2 | 4 | |
2026.05 | 42 | 46 | 30 | 72 | 42 | 2 | 3 |