paper-with-me

Papers

Revisiting Data-Free Knowledge Distillation with Poisoned Teachers

2023-06-04 · Junyuan Hong, Yi Zeng, Shuyang Yu, Lingjuan Lyu, Ruoxi Jia, Jiayu Zhou

Data-free knowledge distillation (KD) helps transfer knowledge from a pre-trained model (known as the teacher model) to a smaller model (known as the student model) without access to the original training data used for training the teacher model. However, the security of the synthetic or out-of-distribution (OOD) data required in data-free KD is largely unknown and under-explored. In this work, we make the first effort to uncover the security risk of data-free KD w.r.t. untrusted pre-trained models. We then propose Anti-Backdoor Data-Free KD (ABD), the first plug-in defensive method for data-free KD methods to mitigate the chance of potential backdoors being transferred. We empirically evaluate the effectiveness of our proposed ABD in diminishing transferred backdoor knowledge while maintaining compatible downstream performances as the vanilla KD. We envision this work as a milestone for alarming and mitigating the potential backdoors in data-free KD. Codes are released at https://github.com/illidanlab/ABD.

📄 PDF Abstract BibTeX arXiv:2306.02368

Code (1)

illidanlab/abd 공식 구현 pytorch

Tasks

Backdoor Defense for Data-Free Distillation with Poisoned TeachersData-free Knowledge DistillationKnowledge Distillation

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

On the Effectiveness of Distillation in Mitigating Backdoors in Pre-trained Encoder

2024-03-06 · Tingxu Han, Shenghan Huang, Ziqi Ding, Weisong Sun 외

In this paper, we study a defense against poisoned encoders in SSL called distillation, which is a defense used in supervised learning originally. Distillation aims to distill knowledge from a given model (a.k.a the teac…

image-classificationImage Classification

Sealing The Backdoor: Unlearning Adversarial Text Triggers In Diffusion Models Using Knowledge Distillation

2025-08-20 · Ashwath Vaithinathan Aravindan, Abha Jha, Matthew Salaway, Atharva Sandeep Bhide 외 arxiv

Text-to-image diffusion models have revolutionized generative AI, but their vulnerability to backdoor attacks poses significant security risks. Adversaries can inject imperceptible textual triggers into training data, ca…

Knowledge Distillation

Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation

2024-10-18 · Shuai Zhao, Xiaobao Wu, Cong-Duy Nguyen, Yanhao Jia 외

Parameter-efficient fine-tuning (PEFT) can bridge the gap between large language models (LLMs) and downstream tasks. However, PEFT has been proven vulnerable to malicious attacks. Research indicates that poisoned LLMs, e…

Backdoor AttackKnowledge DistillationLanguage ModelingLanguage Modelling+3

Revisiting Knowledge Distillation under Distribution Shift

2023-12-25 · Songming Zhang, Ziyu Lyu, Xiaofeng Chen

Knowledge distillation transfers knowledge from large models into small models, and has recently made remarkable achievements. However, few studies has investigated the mechanism of knowledge distillation against distrib…

Data AugmentationDiversityKnowledge Distillation

How to Backdoor the Knowledge Distillation

2025-04-30 · Chen Wu, Qian Ma, Prasenjit Mitra, Sencun Zhu

Knowledge distillation has become a cornerstone in modern machine learning systems, celebrated for its ability to transfer knowledge from a large, complex teacher model to a more efficient student model. Traditionally, t…

Knowledge Distillation