@misc{indiciaef683e37a0fa2,
  title = {From Natural Alignment to Conditional Controllability in Multimodal Dialogue},
  author = {Zeyu Jin and Songtao Zhou and Haoyu Wang and Minghao Tian and Kaifeng Yun and Zhuo Chen and Xiaoyu Qin and Jia Jia},
  year = {2026},
  url = {https://arxiv.org/abs/2603.29162},
  note = {Source identifier: 2603.29162}
}