LLaVA-CMoE: Towards Continual Mixture of Experts for Large Vision-Language Models

@misc{zhao2025llavacmoe,
  title = {LLaVA-CMoE: Towards Continual Mixture of Experts for Large Vision-Language Models},
  author = {Hengyuan Zhao and Ziqin Wang and Qixin Sun and Kaiyou Song and Yilin Li and Xiaolin Hu and Qingpei Guo and Si Liu},
  year = {2025},
  eprint = {2503.21227},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2503.21227},
}