| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Sudoku solving | Sudoku-Extreme (test) | Accuracy87.4 | 6 | |
| Sudoku solving | Sudoku-Extreme (held-out) | Accuracy95.2 | 5 | |
| Sudoku Puzzle Solving | Sudoku-Extreme 17-clue puzzles (test) | Puzzle Accuracy97.3 | 5 | |
| Symbolic Reasoning | Sudoku-Extreme (test) | Accuracy89.34 | 3 |