FOL Theorem Proving on TPTP revised
14.3FLD1 Success RateDREAM
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| DREAMModel=Claude 3.52025.06 | 14.3 | 12.5 | 13.6 | 0 | 0 | 20 | 5.6 | 3 | 22.2 | 10.1 | |
| SubgoalModel=Claude 3.52025.06 | 5.2 | 3.1 | 4.5 | 4.9 | 12.5 | 0 | 0 | 3 | 0 | 3.7 | |
| DREAMModel=DeepSeek-Prover-V2-7B2025.06 | 3.9 | 0 | 11.4 | 9.8 | 12.5 | 30 | 5.6 | 1.5 | 0 | 8.3 | |
| RepeatedModel=DeepSeek-Prover-V2-7B2025.06 | 1.3 | 0 | 6.8 | 7.3 | 0 | 10 | 0 | 1.5 | 11.1 | 4.2 | |
| RepeatedModel=Claude 3.52025.06 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.5 | 0 | 0.2 | |
| SubgoalModel=DeepSeek-Prover-V2-7B2025.06 | 0 | 3.1 | 9.1 | 4.9 | 25 | 20 | 5.6 | 1.5 | 0 | 7.7 |