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

Symbolic Reasoning on CoinFlip

100Calibrated Accuracy

SC+IC (tune)

77.74483.52289.395.078May 29, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.05
100
2024.05
99.9
2024.05
99.9
2024.05
97.7
2024.05
96.5
2024.05
93.9
2024.05
86.2
2024.05
78.6