Mathematical Reasoning (Base-9) on NoRa Inaccurate Rationales
76.7AccuracyCD-CoT
Evaluation Results
| Method | Links | |
|---|---|---|
| CD-CoTModel=Gemini-Pro2024.10 | 76.7 | |
| CD-CoTModel=GPT-3.5-turbo2024.10 | 58.7 | |
| CCModel=Gemini-Pro2024.10 | 43.6 | |
| SCModel=Gemini-Pro2024.10 | 32.3 | |
| CCModel=GPT-3.5-turbo2024.10 | 31.7 | |
| BTModel=Gemini-Pro2024.10 | 26.7 | |
| BaseModel=Gemini-Pro2024.10 | 21.2 | |
| BTModel=GPT-3.5-turbo2024.10 | 18.4 | |
| SCModel=GPT-3.5-turbo2024.10 | 17.3 | |
| CCModel=Mixtral 8x7B2024.10 | 12.5 | |
| BaseModel=GPT-3.5-turbo2024.10 | 10.1 | |
| CD-CoTModel=Mixtral 8x7B2024.10 | 4.7 | |
| BaseModel=Mixtral 8x7B2024.10 | 3.7 | |
| SCModel=LLaMA2-70B2024.10 | 3 | |
| CCModel=LLaMA2-70B2024.10 | 2.8 | |
| BaseModel=LLaMA2-70B2024.10 | 2.7 | |
| CD-CoTModel=LLaMA2-70B2024.10 | 2.7 | |
| SCModel=Mixtral 8x7B2024.10 | 2.7 | |
| BTModel=Mixtral 8x7B2024.10 | 2.4 | |
| BTModel=LLaMA2-70B2024.10 | 0.9 |