1 paper · 1 filter
Oren Sultan, Eitan Stern, Dafna Shahaf
Large language models (LLMs) struggle with formal domains that require rigorous logical deduction and symbolic reasoning, such as mathematical proof generation. We propose a neuro-…