Vision-Centric Activation and Coordination for Multimodal Large Language Models

@misc{wangThu Oct 16 2025 06:38:39 GMT+0000 (Coordinated Universal Time)visioncentricactivationcoordination,
      title={Vision-Centric Activation and Coordination for Multimodal Large Language Models},
      author={Yunnan Wang and Fan Lu and Kecheng Zheng and Ziyuan Huang and Ziqiang Li and Wenjun Zeng and Xin Jin},
      year={Thu Oct 16 2025 06:38:39 GMT+0000 (Coordinated Universal Time)},
      eprint={2510.14349},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2510.14349},
}