paper-with-me

홈 › Papers

Defending Neural Backdoors via Generative Distribution Modeling

2019-10-10 · NeurIPS 2019 12 · Ximing Qiao, Yukun Yang, Hai Li

Neural backdoor attack is emerging as a severe security threat to deep learning, while the capability of existing defense methods is limited, especially for complex backdoor triggers. In the work, we explore the space formed by the pixel values of all possible backdoor triggers. An original trigger used by an attacker to build the backdoored model represents only a point in the space. It then will be generalized into a distribution of valid triggers, all of which can influence the backdoored model. Thus, previous methods that model only one point of the trigger distribution is not sufficient. Getting the entire trigger distribution, e.g., via generative modeling, is a key to effective defense. However, existing generative modeling techniques for image generation are not applicable to the backdoor scenario as the trigger distribution is completely unknown. In this work, we propose max-entropy staircase approximator (MESA), an algorithm for high-dimensional sampling-free generative modeling and use it to recover the trigger distribution. We also develop a defense technique to remove the triggers from the backdoored model. Our experiments on Cifar10/100 dataset demonstrate the effectiveness of MESA in modeling the trigger distribution and the robustness of the proposed defense method.

📄 PDF Abstract BibTeX arXiv:1910.04749

Code (1)

superrrpotato/Defending-Neural-Backdoors-via-Generative-Distribution-Modeling 공식 구현 pytorch

Tasks

Backdoor AttackImage Generationvalid

Similar Papers 제목 키워드 기반

Interactive Simulations of Backdoors in Neural Networks

2024-05-21 · Peter Bajcsy, Maxime Bros

This work addresses the problem of planting and defending cryptographic-based backdoors in artificial intelligence (AI) models. The motivation comes from our lack of understanding and the implications of using cryptograp…

backdoor defense

Oblivious Defense in ML Models: Backdoor Removal without Detection

2024-11-05 · Shafi Goldwasser, Jonathan Shafer, Neekon Vafa, Vinod Vaikuntanathan

As society grows more reliant on machine learning, ensuring the security of machine learning systems against sophisticated attacks becomes a pressing concern. A recent result of Goldwasser, Kim, Vaikuntanathan, and Zamir…

Latent Instruction Representation Alignment: defending against jailbreaks, backdoors and undesired knowledge in LLMs

2026-04-12 · Eric Easley, Sebastian Farquhar arxiv

We address jailbreaks, backdoors, and unlearning for large language models (LLMs). Unlike prior work, which trains LLMs based on their actions when given malign instructions, our method specifically trains the model to c…

DECK: Model Hardening for Defending Pervasive Backdoors

2022-06-18 · Guanhong Tao, Yingqi Liu, Siyuan Cheng, Shengwei An 외

Pervasive backdoors are triggered by dynamic and pervasive input perturbations. They can be intentionally injected by attackers or naturally exist in normally trained models. They have a different nature from the traditi…

Decodermodel

LMSanitator: Defending Prompt-Tuning Against Task-Agnostic Backdoors

2023-08-26 · Chengkun Wei, Wenlong Meng, Zhikun Zhang, Min Chen 외

Prompt-tuning has emerged as an attractive paradigm for deploying large-scale language models due to its strong downstream task performance and efficient multitask serving ability. Despite its wide adoption, we empirical…