conftrace_

trigger feature

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

backdoor attack (2) dataset sanitization (1) poisoned dataset (1) poisoned support set (1) trigger removal (1) imperceptible perturbation (1) trigger optimization (1) support set (1) attack success rate (1) forward process (1)

Papers

DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models AAAI 2024 Does Few-Shot Learning Suffer from Backdoor Attacks? AAAI 2024