Y. Lin, “SEPT: Towards Scalable and Efficient Visual Pre-training”, AAAI, vol. 37, no. 2, pp. 1622–1630, Jun. 2023.