The code is already integrated into the Flash-Linear-Attention repository (https://github.com/fla-org/flash-linear-attention)

🔗 Citation

@article{hu2025comba,
  title={Comba: Improving Nonlinear RNNs with Closed-loop Control},
  author={Hu, Jiaxi and Pan, Yongqi and Du, Jusen and Lan, Disen and Tang, Xiaqiang and Wen, Qingsong and Liang, Yuxuan and Sun, Weigao},
  journal={arXiv preprint arXiv:2506.02475},
  year={2025}
}