Skip to content
conftrace_

safety backdoor

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

backdoor removal (1) harmful content mitigation (1) embedding perturbation (1) backdoor attack (1) bilevel optimization (1) instruction tuning (1) adversarial training (1) adversarial learning (1)

Papers

BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models EMNLP 2024