ReflectRL-Qwen2.5-Math-7B-GRPO

ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning

This repository contains model weights trained using ReflectRL, as presented in the paper ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning.

• Paper: ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning
• Repository: GitHub - bibisbar/ReflectRL

Overview

ReflectRL is a lightweight framework for learning from Golden Negative Trajectories (GNTs) during on-policy post-training. Instead of imitating failed expert trajectories directly, ReflectRL uses them as reflective context during training and gradually transitions the policy back to direct reasoning for inference.

Citation

@article{reflectrl2027,
  title={ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning},
  author={Jinhe Bi and Chennan Zhou and Zengjie Jin and Aniri and Shuo Lu and Wenke Huang and Hu Cao and Xun Xiao and Zhihong Zhu and Volker Tresp and Fei Shen and Yunpu Ma and Tat-Seng Chua},
  year={2027}
}