Data Ranking on Mistake math
0.79AUROCVECFILTER
Evaluation Results
| Method | Links | |
|---|---|---|
| VECFILTERModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.79 | |
| PRISMModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.79 | |
| CONCEPTINFModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.73 | |
| PRISMModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.72 | |
| PRISMModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.71 | |
| PRISMModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.7 | |
| PRISMModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.7 | |
| VECFILTERModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.69 | |
| PRISMModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.69 | |
| CONCEPTINFModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.69 | |
| PRISMModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.69 | |
| PRISMModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.68 | |
| VECFILTERModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.67 | |
| VECFILTERModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.67 | |
| PRISMModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.67 | |
| VECFILTERModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.66 | |
| PRISMModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.66 | |
| VECFILTERModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.66 | |
| CONCEPTINFModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.65 | |
| IF-GUIDEModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.65 | |
| IF-GUIDEModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.63 | |
| PRISMModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.63 | |
| VECFILTERModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.63 | |
| IF-GUIDEModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.63 | |
| PROJDIFFModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.61 | |
| IF-GUIDEModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.61 | |
| IF-GUIDEModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.58 | |
| PROJDIFFModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.57 | |
| VECFILTERModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.56 | |
| VECFILTERModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.56 | |
| IF-GUIDEModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.56 | |
| PROJDIFFModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.56 | |
| IF-GUIDEModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.55 | |
| IF-GUIDEModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.55 | |
| PRISMModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.55 | |
| PROJDIFFModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.54 | |
| IF-GUIDEModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.54 | |
| ICL-EMModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.54 | |
| IF-GUIDEModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.53 | |
| PROJDIFFModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.53 | |
| PROJDIFFModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.53 | |
| ICL-EMModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.52 | |
| PROJDIFFModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.52 | |
| IF-GUIDEModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.52 | |
| ICL-EMModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.51 | |
| PROJDIFFModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.51 | |
| PROJDIFFModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.51 | |
| PROJDIFFModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.51 | |
| ICL-EMModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.51 | |
| IF-GUIDEModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.51 | |
| ICL-EMModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.5 | |
| ICL-EMModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.5 | |
| SAE-FEATModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.49 | |
| VECFILTERModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.48 | |
| SAE-FEATModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.48 | |
| SAE-FEATModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.47 | |
| SAE-FEATModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.47 | |
| ICL-EMModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.47 | |
| ICL-EMModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.46 | |
| CONCEPTINFModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.45 | |
| CONCEPTINFModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.44 | |
| SAE-FEATModel=Llama-3.1-8B-Instruct, Mix ratio=1%2026.05 | 0.41 | |
| SAE-FEATModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.41 | |
| SAE-FEATModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.41 | |
| CONCEPTINFModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.41 | |
| SAE-FEATModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.41 | |
| SAE-FEATModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.41 | |
| CONCEPTINFModel=Qwen-3-8B, Mix ratio=10%2026.05 | 0.4 | |
| CONCEPTINFModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.4 | |
| SAE-FEATModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.4 | |
| ICL-EMModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.4 | |
| SAE-FEATModel=Llama-3.1-8B-Instruct, Mix ratio=10%2026.05 | 0.39 | |
| SAE-FEATModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.39 | |
| CONCEPTINFModel=Qwen-3-8B, Mix ratio=20%2026.05 | 0.38 | |
| CONCEPTINFModel=Qwen-3-8B, Mix ratio=50%2026.05 | 0.37 | |
| ICL-EMModel=Qwen-3-8B, Mix ratio=1%2026.05 | 0.37 | |
| ICL-EMModel=Qwen-3-14B, Mix ratio=10%2026.05 | 0.36 | |
| PROJDIFFModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.35 | |
| PROJDIFFModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.35 | |
| ICL-EMModel=Qwen-3-14B, Mix ratio=1%2026.05 | 0.31 | |
| VECFILTERModel=Llama-3.1-8B-Instruct, Mix ratio=50%2026.05 | 0.29 | |
| VECFILTERModel=Llama-3.1-8B-Instruct, Mix ratio=20%2026.05 | 0.27 | |
| CONCEPTINFModel=Qwen-3-14B, Mix ratio=50%2026.05 | 0.22 | |
| CONCEPTINFModel=Qwen-3-14B, Mix ratio=20%2026.05 | 0.21 |