paper-with-me

홈 › Papers

Reliable Adversarial Distillation with Unreliable Teachers

2021-06-09 · ICLR 2022 4 · Jianing Zhu, Jiangchao Yao, Bo Han, Jingfeng Zhang, Tongliang Liu, Gang Niu, Jingren Zhou, Jianliang Xu, Hongxia Yang

In ordinary distillation, student networks are trained with soft labels (SLs) given by pretrained teacher networks, and students are expected to improve upon teachers since SLs are stronger supervision than the original hard labels. However, when considering adversarial robustness, teachers may become unreliable and adversarial distillation may not work: teachers are pretrained on their own adversarial data, and it is too demanding to require that teachers are also good at every adversarial data queried by students. Therefore, in this paper, we propose reliable introspective adversarial distillation (IAD) where students partially instead of fully trust their teachers. Specifically, IAD distinguishes between three cases given a query of a natural data (ND) and the corresponding adversarial data (AD): (a) if a teacher is good at AD, its SL is fully trusted; (b) if a teacher is good at ND but not AD, its SL is partially trusted and the student also takes its own SL into account; (c) otherwise, the student only relies on its own SL. Experiments demonstrate the effectiveness of IAD for improving upon teachers in terms of adversarial robustness.

📄 PDF Abstract BibTeX arXiv:2106.04928

Code (2)

zfancy/iad 공식 구현 pytorch
boyellow/adaad pytorch

Tasks

Adversarial Robustness

Similar Papers 제목 키워드 기반

Who Taught You That? Tracing Teachers in Model Distillation

2025-02-10 · Somin Wadhwa, Chantal Shaib, Silvio Amir, Byron C. Wallace

Model distillation -- using outputs from a large teacher model to teach a small student model -- is a practical means of creating efficient models for a particular task. We ask: Can we identify a students' teacher based …

Instruction FollowingPOSQuestion Answering

Mitigating Accuracy-Robustness Trade-off via Balanced Multi-Teacher Adversarial Distillation

2023-06-28 · Shiji Zhao, Xizhe Wang, Xingxing Wei

Adversarial Training is a practical approach for improving the robustness of deep neural networks against adversarial attacks. Although bringing reliable robustness, the performance towards clean examples is negatively a…

Adversarial RobustnessKnowledge Distillation

UniSD: Towards a Unified Self-Distillation Framework for Large Language Models

2026-05-07 · Yiqiao Jin, Yiyang Wang, Lucheng Fu, Yijia Xiao 외 arxiv

Self-distillation (SD) offers a promising path for adapting large language models (LLMs) without relying on stronger external teachers. However, SD in autoregressive LLMs remains challenging because self-generated trajec…

Contrastive Learning

Distilling Adversarial Robustness Using Heterogeneous Teachers

2024-02-23 · Jieren Deng, Aaron Palmer, Rigel Mahmood, Ethan Rathbun 외

Achieving resiliency against adversarial attacks is necessary prior to deploying neural network classifiers in domains where misclassification incurs substantial costs, e.g., self-driving cars or medical imaging. Recent …

Adversarial RobustnessKnowledge DistillationSelf-Driving Cars

Learning Better Certified Models from Empirically-Robust Teachers

2026-02-02 · Alessandro De Palma arxiv

Adversarial training attains strong empirical robustness to specific adversarial attacks by training on concrete adversarial perturbations, but it produces neural networks that are not amenable to strong robustness certi…

Knowledge Distillation