Math Reasoning on Math Word Problems (Math WPs) (test)
43.1AccuracyMIDAS
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MIDASNumber of Parameters=8B, Shots=5, Calculator Usage=W calc.2024.09 | 43.1 | 1.844 | |
| MIDASNumber of Parameters=2B, Shots=5, Calculator Usage=W calc.2024.09 | 38.3 | 1.929 | |
| BaselineNumber of Parameters=8B, Shots=5, Calculator Usage=W calc.2024.09 | 34.9 | 1.841 | |
| MIDASNumber of Parameters=8B, Shots=5, Calculator Usage=W/o calc.2024.09 | 32.9 | 1.844 | |
| BaselineNumber of Parameters=8B, Shots=5, Calculator Usage=W/o calc.2024.09 | 27.3 | 1.841 | |
| BaselineNumber of Parameters=2B, Shots=5, Calculator Usage=W calc.2024.09 | 27.1 | 1.926 | |
| MIDASNumber of Parameters=2B, Shots=5, Calculator Usage=W/o calc.2024.09 | 22.5 | 1.929 | |
| BaselineNumber of Parameters=2B, Shots=5, Calculator Usage=W/o calc.2024.09 | 15.4 | 1.926 |