@misc{indiciaed23b63ff596e,
  title = {SceneScaffold: Active Scene-State Construction for Unified 3D Scene Understanding},
  author = {Xiangqi Li and Libo Huang and Jiarui Zhao and Weilun Feng and Chuanguang Yang and Zhulin An and Yongjun Xu},
  year = {2026},
  url = {https://arxiv.org/abs/2609.33518},
  note = {Source identifier: 2609.33518}
}