Combined
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Combined (all envs)
37.1World-State Accuracy (AH)
2
Combined
1.02MRE
2
Combined
86Accuracy
1
Combined (label-stratified)
97.1AUROC
1
Combined AIME'24 AIME'25 HMMT'25 BrUMO'25
0.962Mean Kendall's tau_b Correlation
1
Combined AIME'24 AIME'25 HMMT'25 BrUMO'25
0.865Kendall's tau_b (vs Gold Standard)
1