A Survey of Reinforcement Learning for Large Language Models under Data Scarcity: Challenges and Solutions

Zhiyin Yu, Yuchen Mou, Juncheng Yan +17

Reinforcement learning (RL) has emerged as a powerful post-training paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, reinforcement learni…