@inproceedings{embodiedgptvisionlanguagepretrainingvia,
  title = {EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought},
  author = {Yao Mu and Qinglong Zhang and Mengkang Hu and Wenhai Wang and Mingyu Ding and Jun Jin and Bin Wang and Jifeng Dai and Yu Qiao and Ping Luo},
  year = {2023},
  booktitle = {NeurIPS 2023 11},
  eprint = {2305.15021},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2305.15021v2},
}