@article{latelayerfusionisenoughdualpathvisiontok,
  title = {Late-Layer Fusion is Enough: Dual-Path Vision Token Routing for Multimodal Large Language Models under Visual Saturation},
  author = {Siyuan Liu and Jinyang Wu},
  year = {2026},
  eprint = {2606.09131},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.09131},
}