PulseAugur
中
实时 02:30:49
English(EN) Blackknife: Hard-Label Query-Limited Black-Box Attacks on Heterogeneous Graph Neural Networks

新的Blackknife框架实现了对图神经网络的黑盒攻击

作者 PulseAugur 编辑部 · [1 个来源] ·

研究人员开发了Blackknife，一个旨在对异构图神经网络（HGNNs）执行黑盒对抗性攻击的新颖框架。这种攻击方法在严格的限制下运行，仅需要单跳邻域信息和少量硬标签查询，而无需访问受害者模型的架构、参数、梯度或完整图结构。Blackknife从可观察的邻域构建代理模型，使用连续软权重优化扰动，然后将这些扰动离散化为保留关系的结构重连操作。在ACM、DBLP和IMDB等基准数据集上的实验表明，即使面对基于拓扑的防御策略，Blackknife也能针对各种HGNN模型实现显著的攻击成功率。 AI

影响 凸显了图神经网络的漏洞，可能促使开发新的防御机制以抵御对抗性攻击。

排序理由 这是一篇详细介绍HGNN新攻击框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

新的Blackknife框架实现了对图神经网络的黑盒攻击

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
这是一篇详细介绍HGNN新攻击框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
103 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.LG TIER_1 English(EN) · Honglin Gao, Junhao Ren, Lan Zhao, Yue Yang, Jindong Chang, Gaoxi Xiao ·

    Blackknife：对异构图神经网络的硬标签查询限制黑盒攻击

    arXiv:2606.29240v1 Announce Type: new Abstract: Heterogeneous graph neural networks (HGNNs) have achieved strong performance in modeling complex graph-structured data with multiple node and relation types. However, their robustness under realistic black-box adversarial settings r…

相关实体

相关话题