List your company

Mathematical Reasoning (Base-9) on NoRa Inaccurate Rationales

76.7AccuracyCD-CoT
-2.13218.33438.859.266Oct 31, 2024

Evaluation results

MethodLinks
CD-CoTModel=Gemini-Pro2024.1076.7
CD-CoTModel=GPT-3.5-turbo2024.1058.7
CCModel=Gemini-Pro2024.1043.6
SCModel=Gemini-Pro2024.1032.3
CCModel=GPT-3.5-turbo2024.1031.7
BTModel=Gemini-Pro2024.1026.7
BaseModel=Gemini-Pro2024.1021.2
BTModel=GPT-3.5-turbo2024.1018.4
SCModel=GPT-3.5-turbo2024.1017.3
CCModel=Mixtral 8x7B2024.1012.5
BaseModel=GPT-3.5-turbo2024.1010.1
CD-CoTModel=Mixtral 8x7B2024.104.7
BaseModel=Mixtral 8x7B2024.103.7
SCModel=LLaMA2-70B2024.103
CCModel=LLaMA2-70B2024.102.8
BaseModel=LLaMA2-70B2024.102.7
CD-CoTModel=LLaMA2-70B2024.102.7
SCModel=Mixtral 8x7B2024.102.7
BTModel=Mixtral 8x7B2024.102.4
BTModel=LLaMA2-70B2024.100.9