| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MNIST Addition (N=2) (test) | AS2 | Sum Accuracy99.45 | 7 | 2mo ago | |
| MNIST Addition N=4 (test) | Sum Accuracy99.64 | 5 | 2mo ago | ||
| ADDITIONBase-5 (test) | Best CVX-variant | Accuracy23.5 | 4 | 23d ago | |
| ADDITIONBase-3 (test) | Best CVX-variant | Accuracy36.3 | 4 | 23d ago | |
| ADDITIONBase-2 (test) | Best CVX-variant | Accuracy95.4 | 4 | 23d ago | |
| Arithmetic Addition Base 5, OOD-10x | SG-CVX | Token Accuracy17.2 | 4 | 23d ago | |
| Arithmetic Addition Base 5, OOD-5x | SG-CVX | Token-wise Accuracy17.3 | 4 | 23d ago | |
| Arithmetic Addition Base 5, OOD-2x | SG-CVX | Token-wise Accuracy18.8 | 4 | 23d ago | |
| Arithmetic Addition Base 5, ID n=5 | SG-CVX | Token-wise Accuracy23 | 4 | 23d ago | |
| Arithmetic Addition Base 3 OOD-10x | SG-CVX | Token-wise Accuracy31 | 4 | 23d ago | |
| Arithmetic Addition Base 3 OOD-5x | SG-CVX | Token-wise Accuracy30.7 | 4 | 23d ago | |
| Arithmetic Addition Base 3 OOD-2x | SG-CVX | Token-wise Accuracy31.4 | 4 | 23d ago | |
| Arithmetic Addition Base 3 ID n=5 | CVX | Token-wise Accuracy36.6 | 4 | 23d ago | |
| MNIST Addition N=8 (test) | Digit Accuracy99.98 | 3 | 2mo ago |