List your company

Mathematical Reasoning on GSM8K (Accuracy, Output Tokens, Expense)

84.46AccuracyTALE-EP
26.043241.209156.37571.5409Dec 24, 2024

Evaluation results

MethodLinks
TALE-EPPrompting Strategy=TALE-EP, Model=GPT-4o-mini2024.1284.4677.26279.84
Vanilla CoTPrompting Strategy=Vanilla CoT, Model=GPT-4o-mini2024.1281.35318.1541.09
Directly AnsweringPrompting Strategy=Directly Answering, Model=GPT-4o-mini2024.1228.2912.4639.43