@article{whatdovisualtokensreallyencodeuncovering,
  title = {What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language Models},
  author = {Yingqi Fan and Junlong Tong and Anhao Zhao and Xiaoyu Shen},
  year = {2026},
  eprint = {2603.00510},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2603.00510},
}