Skip to content
conftrace_

feature consistency towards transformation

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

secure training (1) feature consistency (1) poisoned sample (1) adversarial defense (1) backdoor attack (1) deep learning (1) backdoor defense (1) model security (1)

Papers

Effective Backdoor Defense by Exploiting Sensitivity of Poisoned Samples NeurIPS 2022