Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914256286384128 |
|---|---|
| author | Wang, Chuyi Xie, Xiaohui Wang, Tongze Cui, Yong |
| author_facet | Wang, Chuyi Xie, Xiaohui Wang, Tongze Cui, Yong |
| contents | Pre-trained models operating directly on raw bytes have achieved promising performance in encrypted network traffic classification (NTC), but often suffer from shortcut learning-relying on spurious correlations that fail to generalize to real-world data. Existing solutions heavily rely on model-specific interpretation techniques, which lack adaptability and generality across different model architectures and deployment scenarios.
In this paper, we propose BiasSeeker, the first semi-automated framework that is both model-agnostic and data-driven for detecting dataset-specific shortcut features in encrypted traffic. By performing statistical correlation analysis directly on raw binary traffic, BiasSeeker identifies spurious or environment-entangled features that may compromise generalization, independent of any classifier. To address the diverse nature of shortcut features, we introduce a systematic categorization and apply category-specific validation strategies that reduce bias while preserving meaningful information.
We evaluate BiasSeeker on 19 public datasets across three NTC tasks. By emphasizing context-aware feature selection and dataset-specific diagnosis, BiasSeeker offers a novel perspective for understanding and addressing shortcut learning in encrypted network traffic classification, raising awareness that feature selection should be an intentional and scenario-sensitive step prior to model training. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_10180 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification Wang, Chuyi Xie, Xiaohui Wang, Tongze Cui, Yong Machine Learning Networking and Internet Architecture Pre-trained models operating directly on raw bytes have achieved promising performance in encrypted network traffic classification (NTC), but often suffer from shortcut learning-relying on spurious correlations that fail to generalize to real-world data. Existing solutions heavily rely on model-specific interpretation techniques, which lack adaptability and generality across different model architectures and deployment scenarios. In this paper, we propose BiasSeeker, the first semi-automated framework that is both model-agnostic and data-driven for detecting dataset-specific shortcut features in encrypted traffic. By performing statistical correlation analysis directly on raw binary traffic, BiasSeeker identifies spurious or environment-entangled features that may compromise generalization, independent of any classifier. To address the diverse nature of shortcut features, we introduce a systematic categorization and apply category-specific validation strategies that reduce bias while preserving meaningful information. We evaluate BiasSeeker on 19 public datasets across three NTC tasks. By emphasizing context-aware feature selection and dataset-specific diagnosis, BiasSeeker offers a novel perspective for understanding and addressing shortcut learning in encrypted network traffic classification, raising awareness that feature selection should be an intentional and scenario-sensitive step prior to model training. |
| title | Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification |
| topic | Machine Learning Networking and Internet Architecture |
| url | https://arxiv.org/abs/2601.10180 |