analiza aiqbench
This research introduces Latent Sculpting, a novel framework that enhances intrusion detection systems by explicitly structuring latent representations to handle out-of-distribution attacks without retraining. By decoupling geometry learning from density modeling using Transformers and Masked Autoregressive Flows, the method achieves high zero-shot performance on unseen cyber threats like stealthy infiltrations and low-volume DoS variants. The approach offers a robust solution for securing networks against previously unknown attack patterns where traditional models typically fail.
Latent Sculpting for Zero-Shot Generalization: A Manifold Learning Approach to Out-of-Distribution Anomaly Detection
arXiv:2512.22179v3 Announce Type: replace Abstract: Detecting previously unseen attacks remains a major challenge for machine learning-based intrusion detection systems. Deep models trained on network traffic often achieve high accuracy on known attacks but fail under distributional shift because their decision boundaries are tightly coupled to the training data distribution. We introduce Latent Sculpting, a two-stage anomaly detection framework that improves robustness by explicitly structuring the latent representation before density estimation. The first stage trains a Transformer-based tabular encoder using a novel Binary Latent Sculpting loss, which encourages benign traffic to form a compact latent cluster while enforcing separation from anomalous patterns. The second stage fits a Masked Autoregressive Flow to the resulting latent space to produce calibrated probabilistic anomaly scores. Under a strict zero-shot evaluation protocol on the CIC-IDS-2017 benchmark, Stage 1 attains an F1-score of 0.98 on known attacks, while Stage 2 -- evaluated at the balanced threshold (85th-percentile) -- achieves a zero-shot OOD F1-score of 0.867 and AUROC of 0.913. The model successfully detects difficult distribution shifts including stealthy infiltration attacks (78.7% recall, peaking at 97.2%) and low-volume DoS variants (>94% recall), scenarios where conventional approaches often fail. Our results suggest that explicitly separating latent geometry learning from density modeling provides a stable approach for detecting zero-day cyber threats.
Abstract:Detecting previously unseen attacks remains a major challenge for machine learning-based intrusion detection systems. Deep models trained on network traffic often achieve high accuracy on known attacks but fail under distributional shift because their decision boundaries are tightly coupled to the training data distribution. We introduce Latent Sculpting, a two-stage anomaly detection framework that improves robustness by explicitly structuring the latent representation before density estimation. The first stage trains a Transformer-based tabular encoder using a novel Binary Latent Sculpting loss, which encourages benign traffic to form a compact latent cluster while enforcing separation from anomalous patterns. The second stage fits a Masked Autoregressive Flow to the resulting latent space to produce calibrated probabilistic anomaly scores. Under a strict zero-shot evaluation protocol on the CIC-IDS-2017 benchmark, Stage 1 attains an F1-score of 0.98 on known attacks, while Stage 2 -- evaluated at the balanced threshold (85th-percentile) -- achieves a zero-shot OOD F1-score of 0.867 and AUROC of 0.913. The model successfully detects difficult distribution shifts including stealthy infiltration attacks (78.7% recall, peaking at 97.2%) and low-volume DoS variants (>94% recall), scenarios where conventional approaches often fail. Our results suggest that explicitly separating latent geometry learning from density modeling provides a stable approach for detecting zero-day cyber threats.
6 pages, 0 figures. Accepted for publication in the 35th International Conference on Computer Communications and Networks (ICCCN 2026). Code available at: this https URL
Machine Learning (cs.LG); Cryptography and Security (cs.CR)
arXiv:2512.22179 [cs.LG]
(or arXiv:2512.22179v3 [cs.LG] for this version)
Text integral sub licența deschisă a publicației · 246 cuvinte
Subiecte conexe
REALM: An RGB- and Event-Aligned Latent Manifold for Cross-Modal Perception
arXiv · 22 sept. 2026
MAGNET: Manifold-Aware Graph Diffusion Network for Connectome Generation
arXiv · 15 oct. 2026
AST: Adaptive, Seamless, and Training-Free Precise Speech Editing
arXiv · 5 aug. 2026
[AINews] not much happened today
Latent Space · 3 oct. 2026