Automated Theorem Proving on miniF2F-test
Metric: Pass@64 (higher is better)
Metric: Pass@64 (higher is better)
| # | Model↕ | Pass@64▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | DeepSeek-Prover-V1.5 | 50.7 | Yes | DeepSeek-Prover-V1.5: Harnessing Proof Assistant... | 2024-08-15 | Code |
| 2 | DeepSeek-Prover | 46.3 | Yes | DeepSeek-Prover: Advancing Theorem Proving in LL... | 2024-05-23 | - |
| 3 | Evariste | 41 | Yes | HyperTree Proof Search for Neural Theorem Proving | 2022-05-23 | - |
| 4 | Evariste-7d | 40.6 | No | HyperTree Proof Search for Neural Theorem Proving | 2022-05-23 | - |
| 5 | Evariste-1d | 38.9 | No | HyperTree Proof Search for Neural Theorem Proving | 2022-05-23 | - |
| 6 | Lean Expert Iteration | 36.6 | Yes | Formal Mathematics Statement Curriculum Learning | 2022-02-03 | Code |
| 7 | GPT-f | 36.6 | No | HyperTree Proof Search for Neural Theorem Proving | 2022-05-23 | - |