Source code for our paper:

CARFT: Boosting LLM Reasoning via Contrastive Learning with Annotated Chain-of-Thought-based Reinforced Fine-Tuning

please use cl_trainer_v4 to reproduce the results in the paper.

https://mp.weixin.qq.com/s/E7H3DCri6_xmBLwUv485Zw

Feel free to contact me if there is any question:

zhuwnq@outlook.com