Physical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0Cliffhanger
5
Jun 30, 2026
97.2mIoU
5
Apr 27, 2026
24.3AUCCESS
5
Feb 26, 2026
77.6AUCCESS
5
Feb 26, 2026
56.31AUCCESS
5
Feb 26, 2026
85.49AUCCESS
5
Feb 26, 2026
34Accuracy
4
Jun 30, 2026
72Accuracy
4
Jun 30, 2026
0.436Comprehensive Score
4
Apr 10, 2026
23.5Spatial IoU
3
Apr 10, 2026
54Accuracy
2
Jun 30, 2026