Language Understanding on MMLU CF
74.5ScoreCAP-CoT
Evaluation Results
| Method | Links | |
|---|---|---|
| CAP-CoTBackbone=GPT-4o2026.04 | 74.5 | |
| ToTBackbone=GPT-4o2026.04 | 73.5 | |
| ECONBackbone=GPT-4o2026.04 | 73.5 | |
| MADBackbone=GPT-4o2026.04 | 73 | |
| AoTBackbone=GPT-4o2026.04 | 73 | |
| CoT-SCBackbone=GPT-4o2026.04 | 72.8 | |
| CAP-CoTBackbone=GPT-4o-mini2026.04 | 72.5 | |
| FoTBackbone=GPT-4o2026.04 | 72.4 | |
| PromptAgentBackbone=GPT-4o2026.04 | 72.3 | |
| CAP-CoTBackbone=Deepseek-V32026.04 | 72.2 | |
| GoTBackbone=GPT-4o2026.04 | 72.2 | |
| CCoTBackbone=GPT-4o2026.04 | 72.1 | |
| ToTBackbone=Deepseek-V32026.04 | 72 | |
| ECONBackbone=Deepseek-V32026.04 | 71.8 | |
| Self-RefineBackbone=GPT-4o2026.04 | 71.8 | |
| GoTBackbone=GPT-4o-mini2026.04 | 71.6 | |
| CAP-CoTBackbone=Qwen-turbo2026.04 | 71.6 | |
| ECONBackbone=GPT-4o-mini2026.04 | 71.5 | |
| CoT-SCBackbone=Deepseek-V32026.04 | 71.5 | |
| CoTBackbone=GPT-4o2026.04 | 71.5 | |
| ToTBackbone=Qwen-turbo2026.04 | 71.3 | |
| ECONBackbone=Qwen-turbo2026.04 | 71.2 | |
| MADBackbone=Deepseek-V32026.04 | 71.2 | |
| AFlowBackbone=GPT-4o2026.04 | 71.2 | |
| CoT-SCBackbone=GPT-4o-mini2026.04 | 71.1 | |
| MADBackbone=GPT-4o-mini2026.04 | 71 | |
| AoTBackbone=GPT-4o-mini2026.04 | 70.9 | |
| CoT-SCBackbone=Qwen-turbo2026.04 | 70.8 | |
| MADBackbone=Qwen-turbo2026.04 | 70.8 | |
| AoTBackbone=Deepseek-V32026.04 | 70.8 | |
| FoTBackbone=GPT-4o-mini2026.04 | 70.6 | |
| PromptAgentBackbone=Deepseek-V32026.04 | 70.6 | |
| PromptAgentBackbone=GPT-4o-mini2026.04 | 70.5 | |
| AoTBackbone=Qwen-turbo2026.04 | 70.5 | |
| FoTBackbone=Deepseek-V32026.04 | 70.5 | |
| CCoTBackbone=GPT-4o-mini2026.04 | 70.4 | |
| FoTBackbone=Qwen-turbo2026.04 | 70.4 | |
| GoTBackbone=Deepseek-V32026.04 | 70.3 | |
| ToTBackbone=GPT-4o-mini2026.04 | 70.2 | |
| PromptAgentBackbone=Qwen-turbo2026.04 | 70.1 | |
| CCoTBackbone=Qwen-turbo2026.04 | 70.1 | |
| GoTBackbone=Qwen-turbo2026.04 | 70.1 | |
| Self-RefineBackbone=Deepseek-V32026.04 | 70.1 | |
| CCoTBackbone=Deepseek-V32026.04 | 70.1 | |
| CoTBackbone=Deepseek-V32026.04 | 69.9 | |
| GHS-TDABackbone=Qwen 2-14B2026.02 | 69.8 | |
| AFlowBackbone=Deepseek-V32026.04 | 69.8 | |
| Self-RefineBackbone=GPT-4o-mini2026.04 | 69.7 | |
| CoTBackbone=GPT-4o-mini2026.04 | 69.6 | |
| AFlowBackbone=GPT-4o-mini2026.04 | 69.5 | |
| Self-RefineBackbone=Qwen-turbo2026.04 | 69.5 | |
| CoTBackbone=Qwen-turbo2026.04 | 69.4 | |
| AFlowBackbone=Qwen-turbo2026.04 | 69.3 | |
| AoTBackbone=Qwen 2-14B2026.02 | 68.9 | |
| GoTBackbone=Qwen 2-14B2026.02 | 67.8 | |
| ToTBackbone=Qwen 2-14B2026.02 | 67.6 | |
| APBackbone=GPT-4o2026.04 | 67.5 | |
| GHS-TDABackbone=Llama 3-8B2026.02 | 67.3 | |
| AoTBackbone=Llama 3-8B2026.02 | 66.65 | |
| APBackbone=Deepseek-V32026.04 | 66.1 | |
| GoTBackbone=Llama 3-8B2026.02 | 65.99 | |
| APBackbone=GPT-4o-mini2026.04 | 65.8 | |
| ToTBackbone=Llama 3-8B2026.02 | 65.71 | |
| CoTBackbone=Llama 3-8B2026.02 | 65.42 | |
| APBackbone=Qwen-turbo2026.04 | 65.2 | |
| CoTBackbone=Qwen 2-14B2026.02 | 65.1 |