1 paper
Yanrui Wu, Lingling Zhang, Xinyu Zhang +5
Evaluations of large language models (LLMs) primarily emphasize convergent logical reasoning, where success is defined by producing a single correct proof. However, many real-world…