Obliviate: Neutralizing Task-agnostic Backdoors within the Parameter-efficient Fine-tuning Paradigm

TL;DR

Obliviate is a novel defense method for parameter-efficient fine-tuning of large language models that effectively neutralizes task-agnostic backdoors by amplifying benign neurons and penalizing trigger tokens, enhancing security.

Contribution

This paper introduces Obliviate, the first PEFT-compatible defense that significantly reduces backdoor success rates and defends against various attack types.

Findings

Reduces attack success rate of backdoors by 83.6%

Effective against task-specific and adaptive backdoor attacks

Compatible with major PEFT architectures

Abstract

Parameter-efficient fine-tuning (PEFT) has become a key training strategy for large language models. However, its reliance on fewer trainable parameters poses security risks, such as task-agnostic backdoors. Despite their severe impact on a wide range of tasks, there is no practical defense solution available that effectively counters task-agnostic backdoors within the context of PEFT. In this study, we introduce Obliviate, a PEFT-integrable backdoor defense. We develop two techniques aimed at amplifying benign neurons within PEFT layers and penalizing the influence of trigger tokens. Our evaluations across three major PEFT architectures show that our method can significantly reduce the attack success rate of the state-of-the-art task-agnostic backdoors (83.6%$\downarrow$). Furthermore, our method exhibits robust defense capabilities against both task-specific backdoors and adaptive attacks. Source code will be obtained at https://github.com/obliviateARR/Obliviate.