Attack Smarter: Attention-Driven Fine-Grained Webpage Fingerprinting Attacks

2025-06-24

AI Summary

Web fingerprinting (WF) faces two key challenges in large-scale multi-label browsing scenarios: high similarity of traffic features across subpages, and trajectory-level feature mixing—where a single trace contains multiple labels with non-aligned, position-variant features. To address these, this paper proposes ADWPF, a fine-grained web fingerprinting attack framework. Methodologically, ADWPF introduces an attention-driven data augmentation strategy—comprising attention-guided cropping and masking—and integrates a residual attention mechanism to model temporal-sensitive, class-specific representations. Additionally, a self-attention module is employed to extract low-dimensional features from both original and augmented traffic, effectively capturing global contextual dependencies. Experiments demonstrate that ADWPF consistently outperforms state-of-the-art methods across multiple benchmark datasets. Notably, it achieves the first successful fine-grained subpage identification under realistic conditions involving multi-label annotations, high label overlap, and non-aligned traffic traces.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Application Domains: Mobility, Driving & Flight
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Website Fingerprinting (WF) attacks aim to infer which websites a user is visiting by analyzing traffic patterns, thereby compromising user anonymity. Although this technique has been demonstrated to be effective in controlled experimental environments, it remains largely limited to small-scale scenarios, typically restricted to recognizing website homepages. In practical settings, however, users frequently access multiple subpages in rapid succession, often before previous content fully loads. WebPage Fingerprinting (WPF) generalizes the WF framework to large-scale environments by modeling subpages of the same site as distinct classes. These pages often share similar page elements, resulting in lower inter-class variance in traffic features. Furthermore, we consider multi-tab browsing scenarios, in which a single trace encompasses multiple categories of webpages. This leads to overlapping traffic segments, and similar features may appear in different positions within the traffic, thereby increasing the difficulty of classification. To address these challenges, we propose an attention-driven fine-grained WPF attack, named ADWPF. Specifically, during the training phase, we apply targeted augmentation to salient regions of the traffic based on attention maps, including attention cropping and attention masking. ADWPF then extracts low-dimensional features from both the original and augmented traffic and applies self-attention modules to capture the global contextual patterns of the trace. Finally, to handle the multi-tab scenario, we employ the residual attention to generate class-specific representations of webpages occurring at different temporal positions. Extensive experiments demonstrate that the proposed method consistently surpasses state-of-the-art baselines across datasets of different scales.

Problem

• Infer visited websites via traffic patterns, risking anonymity
• Handle subpage recognition in large-scale browsing scenarios
• Address multi-tab browsing with overlapping traffic segments

Innovation

• Attention-driven fine-grained WPF attack
• Targeted augmentation using attention maps
• Residual attention for multi-tab scenarios