Model Card for IF-CRITIC-14B

Model Description

• Model type: Generative reward model
• Language(s) (NLP): Chinese
• License: MIT
• Finetuned from model: Qwen/Qwen2.5-14B-Instruct

Model Sources

• Repository: https://github.com/thu-coai/IF-CRITIC.
• Paper: https://arxiv.org/pdf/2511.01014
• Please refer to our paper and our GitHub repository for more details.

Citation

If this model helps, please kindly cite us:

@article{wen2025if,
 title={IF-CRITIC: Towards a Fine-Grained LLM Critic for Instruction-Following Evaluation},
 author={Wen, Bosi and Niu, Yilin and Wang, Cunxiang and Ke, Pei and Ling, Xiaoying and Zhang, Ying and Zeng, Aohan and Wang, Hongning and Huang, Minlie},
 journal={arXiv preprint arXiv:2511.01014},
 year={2025}
}

Base model