@inproceedings{segagentexploringpixelunderstanding1,
  title = {SegAgent: Exploring Pixel Understanding Capabilities in MLLMs by Imitating Human Annotator Trajectories},
  author = {Muzhi Zhu and Yuzhuo Tian and Hao Chen and Chunluan Zhou and Qingpei Guo and Yang Liu and Ming Yang and Chunhua Shen},
  year = {2025},
  booktitle = {CVPR 2025 1},
  eprint = {2503.08625},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2503.08625v1},
}