@misc{indiciaee4ea63f28c03,
  title = {Information-Theoretic Constraints for Continual Vision-Language-Action Alignment},
  author = {Libang Zhao and Qixin Zeng and Hongyin Zhang and Donglin Wang},
  year = {2026},
  url = {https://arxiv.org/abs/2603.13335},
  note = {Source identifier: 2603.13335}
}