starriver030515/Qwen2.5-Math-1.5B-16k

The base Qwen2.5-Math-1.5B model used by HAPO. We change to rope_theta from 10000 to 40000 and extend the context window to 16k. Also, we modify the chat_template for the system prompt and add .

Citation

If you find our model, data, or evaluation code useful, please kindly cite our paper:

@misc{liu2025uniformheterogeneoustailoringpolicy,
      title={From Uniform to Heterogeneous: Tailoring Policy Optimization to Every Token's Nature}, 
      author={Zheng Liu and Mengjie Liu and Siwei Wen and Mengzhang Cai and Bin Cui and Conghui He and Wentao Zhang},
      year={2025},
      eprint={2509.16591},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2509.16591}, 
}