List your company

Reasoning on AQUA, GSM8K, SVAMP, ASDiv, StrategyQA, CSQA, ARC, LastLetter (test)

72.84AQUA AccuracySGE
43.012850.756458.566.2436May 28, 2024

Evaluation results

MethodLinks
SGEBackbone=GPT-3.5, Use Code Interpreter=true2024.0572.8486.1886.4490.1777.4479.0987.8378.7882.27
Decomp PromptingBackbone=GPT-3.5, Use Code Interpreter=true2024.0571.6482.1184.3489.8376.1678.8487.4877.8981.11
Refine PromptingBackbone=GPT-3.5, Use Code Interpreter=true2024.0558.8176.180.4487.3573.3874.1385.7174.7876.34
CoT PromptingBackbone=GPT-3.5, Use Code Interpreter=true2024.0558.3575.8480.387.3972.9774.3685.4874.6376.17
IO PromptingBackbone=GPT-3.5, Use Code Interpreter=true2024.0544.1657.0560.0265.4354.755.6864.3556.2157.2