Math Word Problem solving on Math23k Chinese (test)
86.5AccuracyATHENA
Evaluation Results
| Method | Links | |
|---|---|---|
| ATHENABackbone=RoBERTa-large2023.11 | 86.5 | |
| DeductReasonerBackbone=RoBERTa-large2023.11 | 85.8 | |
| DeductReasonerBackbone=RoBERTa-base2023.11 | 85.1 | |
| ATHENABackbone=RoBERTa-base2023.11 | 84.4 | |
| Graph-to-TreeBackbone=Random embedding2023.11 | 77.4 | |
| R-TransformerBackbone=RoBERTa-base2023.11 | 76.9 | |
| GTSBackbone=Random embedding2023.11 | 75.6 | |
| TransformerBackbone=Random embedding2023.11 | 61.5 |