Automated Theorem Proving On Metamath Setmm
评估指标
Percentage correct
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Percentage correct | Paper Title | Repository |
---|---|---|---|
GPT-f | 56.2 | Generative Language Modeling for Automated Theorem Proving | - |
Evariste | - | HyperTree Proof Search for Neural Theorem Proving | - |
MetaGen-IL + Holophrasm | 22.1 | Learning to Prove Theorems by Learning to Generate Theorems | |
Holophrasm | 14.3 | Holophrasm: a neural Automated Theorem Prover for higher-order logic |
0 of 4 row(s) selected.