1 paper
Kai Hua, Steven Wu, Ge Zhang +1
Recently, there has been growing interest in collecting reasoning-intensive pretraining data to improve LLMs' complex reasoning ability. Prior approaches typically rely on supervis…