Math Word Problem solving on UnbiasedMWP 1:N Chinese (test)
48.4AccuracyATHENA
Evaluation Results
| Method | Links | |
|---|---|---|
| ATHENABackbone=RoBERTa-large2023.11 | 48.4 | |
| ATHENABackbone=RoBERTa-base2023.11 | 35.4 | |
| DeductReasonerBackbone=RoBERTa-large2023.11 | 33.7 | |
| DeductReasonerBackbone=RoBERTa-base2023.11 | 26.5 | |
| Graph-to-TreeBackbone=Random embedding2023.11 | 24.3 | |
| GTSBackbone=Random embedding2023.11 | 22.8 | |
| TransformerBackbone=Random embedding2023.11 | 16.9 | |
| R-TransformerBackbone=RoBERTa-base2023.11 | 14.9 |