Math Word Problem solving on UnbiasedMWP Chinese (test)
42AccuracyATHENA
Evaluation Results
| Method | Links | |
|---|---|---|
| ATHENABackbone=RoBERTa-large2023.11 | 42 | |
| ATHENABackbone=RoBERTa-base2023.11 | 36.2 | |
| DeductReasonerBackbone=RoBERTa-large2023.11 | 34.9 | |
| DeductReasonerBackbone=RoBERTa-base2023.11 | 31.6 | |
| Graph-to-TreeBackbone=Random embedding2023.11 | 27.2 | |
| GTSBackbone=Random embedding2023.11 | 26.2 | |
| TransformerBackbone=Random embedding2023.11 | 20.5 | |
| R-TransformerBackbone=RoBERTa-base2023.11 | 18.3 |