@inproceedings{ijcai2026p666,
  title     = {MathCritique: Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervision},
  author    = {Xi, Zhiheng and Yang, Dingwen and Huang, Jixuan and Tang, Jiafu and Guo, Xin and Li, Guanyu and Ding, Yiwen and He, Wei and Hong, Boyang and Dou, Shihan and Zhan, WenYu and Wang, Xiao and Shi, Xiaowei and Zhai, Yitao and Weng, Rongxiang and Wang, Jingang and Zheng, Rui and Ji, Tao and Gui, Tao and Wu, Zuxuan and Zhang, Qi and Qiu, Xipeng and Huang, Xuanjing and Jiang, Yu-Gang},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {5982--5990},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/666},
  url       = {https://doi.org/10.24963/ijcai.2026/666},
}