@misc{indiciae559b27a1026a,
  title = {ID-Align: RoPE-Conscious Position Remapping for Dynamic High-Resolution Adaptation in Vision-Language Models},
  author = {Bozhou Li and Wentao Zhang},
  year = {2026},
  url = {https://arxiv.org/abs/2505.21465},
  note = {Source identifier: 2505.21465}
}