Formal Proof Refactoring on PutnamBench 96 problems
4.19Average ScoreProof-Refactor
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Proof-Refactor2026.06 | 4.19 | 4.21 | 4.14 | 4.19 | 4.24 | 4.18 | 6.5 | 118.1 | |
| Baseline2026.06 | 3.75 | 3.8 | 3.46 | 3.97 | 3.84 | 3.67 | 4 | 105.5 |
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Proof-Refactor2026.06 | 4.19 | 4.21 | 4.14 | 4.19 | 4.24 | 4.18 | 6.5 | 118.1 | |
| Baseline2026.06 | 3.75 | 3.8 | 3.46 | 3.97 | 3.84 | 3.67 | 4 | 105.5 |