Using mixup as a regularizer can surprisingly improve accuracy and out-of-distribution robustness by Pinto, F, Yang, H, Lim, SN, Torr, PHS, Dokania, PK

A continuous mapping for augmentation design by Tian, K, Lin, C, Lim, SN, Ouyang, W, Dokania, PK, Torr, P

Computationally budgeted continual learning: what does matter? by Prabhu, A, Hammoud, HAAK, Dokania, P, Torr, P, Lim, SN, Ghanem, B, Bibi, A

uCAP: an unsupervised prompting method for vision-language models by Nguyen, AT, Tai, KS, Chen, BC, Shukla, SN, Yu, H, Torr, P, Tian, TP, Lim, SN