List your company

Question Answering and Reasoning on Macro-average (MMLU, MATH, GSM8K, BBH)

46Cost ReductionCISC
8.5618.282837.72Feb 10, 2025

Evaluation results

MethodLinks
CISCConfidence Method=P(True), Budget=102025.02461.1
CISCConfidence Method=P(True), Budget=52025.02411.6
CISCConfidence Method=Response Probability, Budget=102025.02310.8
CISCConfidence Method=Verbal 1-100, Budget=102025.02300.4
CISCConfidence Method=Verbal 1-100, Budget=52025.02220.8
CISCConfidence Method=Response Probability, Budget=52025.02221.1
CISCConfidence Method=Verbal Binary, Budget=52025.02180.4
CISCConfidence Method=Verbal Binary, Budget=102025.02100.2