Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Meta-reasoning quality assessment on APPS

85.6Thoroughness

MCTS-Judge

11.55230.7765069.224Feb 18, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
85.677.384.153
2025.02
74.259.183.353.8
2025.02
7274.261.450.8
2025.02
71.250.865.247.7
2025.02
68.956.859.956.1
2025.02
64.448.566.745.5
2025.02
35.651.533.354.5
2025.02
31.143.240.143.9
2025.02
28.849.234.852.3
2025.02
2825.838.649.2
2025.02
25.840.916.746.2
2025.02
14.422.715.947