Research map / Adversarial robustness and security / Backdoor and poisoning
Disentangled Dynamic Intrusion Detection
Chenyang Qiu, Guoshun Nan, Hongrui Xia, Zheng Weng, Xueting Wang, Meng Shen, Xiaofeng Tao, Jun Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add statistical disentanglement (SD) (ablation) | DIDS · CIC-ToN-IoT binary | F1: 91.69 -> 97.92 | Improved |
| add representational disentanglement (RD) (ablation) | DIDS · CIC-ToN-IoT binary | F1: 89.15 -> 97.92 | Improved |
| add multi-layer graph diffusion (ML-GRAND) (ablation) | DIDS · CIC-ToN-IoT binary | AUC: 79.32 -> 96.04 | Improved |
| add multi-scale latent optimization space (LOS) (ablation) | MFL · CIC-ToN-IoT few-shot | F1: 75.82 -> 97.47 | Improved |
| add disentanglement regularization (DR) (ablation) | MFL · CIC-ToN-IoT few-shot | F1: 74.90 -> 97.47 | Improved |
| add both LOS and DR to naive self-expressiveness base (ablation) | SE · CIC-ToN-IoT few-shot | F1: 87.80 -> 97.47 | Improved |
| use only one of LOS or DR instead of both (ablation) | MFL · CIC-ToN-IoT few-shot | F1: 87.80 -> 75.82 | Hurt |
| add MFL to DIDS for multi-classification | DIDS · CIC-ToN-IoT multi-class | F1: 83.56 -> 96.78 | Improved |
| add MFL to DIDS for multi-classification | DIDS · CIC-BoT-IoT multi-class | F1: 82.98 -> 85.72 | Improved |
| add MFL to DIDS for multi-classification | DIDS · NF-UNSW-NB15-v2 multi-class | F1: 93.55 -> 95.71 | Improved |
| use DIDS dynamic graph over static E-GraphSAGE for unknown MITM | E-GraphSAGE · CIC-ToN-IoT unknown MITM | F1: 6.05 -> 34.91 | Improved |
| use DIDS-MFL instead of an LLM | LLM · CIC-ToN-IoT 100 samples | F1: 66.67% -> 99.52% | Improved |
Closest papers by meaning
- DiffuPac: Contextual Mimicry in Adversarial Packets Generation via Diffusion Model (NeurIPS 2024 · similarity 0.86)
- Revolutionizing Encrypted Traffic Classification with MH-Net: A Multi-View Heterogeneous Graph Model (AAAI 2025 · similarity 0.84)
- MIETT: Multi-Instance Encrypted Traffic Transformer for Encrypted Traffic Classification (AAAI 2025 · similarity 0.83)
- Self-Supervised Learning of Graph Representations for Network Intrusion Detection (NeurIPS 2025 · similarity 0.83)
- MalDetectFormer: Leveraging Sparse SpatioTemporal Information for Effective Malicious Traffic Detection (AAAI 2025 · similarity 0.82)
- Dissect Black Box: Interpreting for Rule-Based Explanations in Unsupervised Anomaly Detection (NeurIPS 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
