1 paper
Xiwei Xuan, Ziquan Deng, Hsuan-Tien Lin +1
Recent studies highlight that deep learning models often learn spurious features mistakenly linked to labels, compromising their reliability in real-world scenarios where such corr…