1 paper
Leizhen Zhang, Shuhan Chen, Sheng Chen
Large language models (LLMs) are increasingly used for tasks that implicitly reduce to Boolean satisfiability (SAT), yet their reasoning ability on SAT remains unclear. We present…