List your company

Mathematical Reasoning on MATH Full (test)

0.539Overall AccuracyComplex CoT
0.420440.451220.4820.51278Aug 8, 2023

Evaluation results

MethodLinks
Complex CoTw/ PHP=true, Number of shots=8-shot, Backbone=GPT-4, Code Environment=None2023.080.5390.2980.4190.5570.5630.7380.7432.8494
Complex CoTw/ PHP=false, Number of shots=8-shot, Backbone=GPT-4, Code Environment=None2023.080.50360.2670.3650.4960.5310.7160.708—
CoTw/ PHP=false, Backbone=GPT-4, Code Environment=None2023.080.425———————