Mathematical Question Answering on Math23K
Metric: Accuracy (training-test) (higher is better)
Metric: Accuracy (training-test) (higher is better)
| # | Model↕ | Accuracy (training-test)▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | Multi-view* (ours) | 87.1 | Yes | Multi-View Reasoning: Consistent Contrastive Lea... | 2022-10-21 | Code |
| 2 | ATHENA (roberta-large) | 86.5 | No | ATHENA: Mathematical Reasoning with Thought Expa... | 2023-11-02 | Code |
| 3 | Exp-Tree | 86.2 | No | An Expression Tree Decoding Strategy for Mathema... | 2023-10-14 | Code |
| 4 | Generate and Rank | 85.4 | No | Generate & Rank: A Multi-task Framework for Math... | 2021-09-07 | - |
| 5 | REAL2: Memory-augmented Solver | 85.2 | No | - | - | - |
| 6 | MWP-BERT | 84.7 | No | MWP-BERT: Numeracy-Augmented Pre-training for Ma... | 2021-07-28 | Code |
| 7 | ATHENA (roberta-base) | 84.4 | No | ATHENA: Mathematical Reasoning with Thought Expa... | 2023-11-02 | Code |
| 8 | Recall and Learn | 82.3 | No | Recall and Learn: A Memory-augmented Solver for ... | 2021-09-27 | Code |
| 9 | Graph2Tree | 77.4 | No | - | - | Code |
| 10 | GROUP-ATT | 69.5 | No | - | - | Code |
| 11 | T-RNN | 66.9 | No | - | - | - |