| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Isabelle (IS) Premise Selection (test) | Llama-3.1 (RULES) | Exact Match Accuracy86.3 | 12 | 3mo ago | |
| HolStep (test) | FormulaNet | Accuracy90.3 | 8 | 3mo ago | |
| LeanDojo Benchmark 4 Lean 3 (novel_premises) | ReProver | R@19.8 | 6 | 3mo ago | |
| LeanDojo Benchmark (random) | ReProver (Ours) | R@113.5 | 5 | 3mo ago | |
| Dl (test) | GPT | Accuracy83 | 4 | 28d ago | |
| Ds (test) | T5 | Accuracy99 | 4 | 28d ago | |
| D (test) | T5 | Accuracy94 | 2 | 28d ago | |
| Mizar (test) | BidirDagLSTM-AttDagLSTM | Accuracy81 | 2 | 3mo ago | |
| LeanDojo Benchmark 4 Lean 3 (random) | ReProver | R@112.8 | 1 | 3mo ago |