XSec is introduced, a self-explainable deep architecture developed for security applications that produces deterministic explanations for a fixed trained model and input and substantially reduces explanation latency compared with approximation-based and perturbation-based post-hoc methods.
Abstract
Deep learning models have become integral to security applications due to their ability to model complex relationships in data and detect sophisticated threats. However, their complexity makes it difficult to understand how predictions are generated, posing significant challenges for interpretability, particularly in security applications where transparency is critical. Existing explanation methods, such as visual explanation techniques and post-hoc approaches, suffer from several limitations: reduced faithfulness due to local approximation errors, instability caused by reliance on randomness, and computational inefficiency that hinders real-time usage. To address these issues, we introduce XSec, a self-explainable deep architecture developed for security applications. During training, XSec uses a novel mask-based approach to extract informative sub-features from the data and learns prototypes, representative patterns that characterize each class. XSec then leverages the prototypes in a dedicated similarity layer at test time to compute similarity scores and generates interpretable explanations without the need for post-hoc analysis. We evaluate XSec across five diverse security scenarios, demonstrating its ability to achieve an average classification accuracy of 97.33% with minimal performance compromise. XSec produces deterministic explanations for a fixed trained model and input and substantially reduces explanation latency compared with approximation-based and perturbation-based post-hoc methods. Through this effort, we extend the applicability of self-explainable AI to security applications, bridging the gap between deep learning performance and the need for explainability in critical scenarios.
Sparse Autoencoders (SAEs) offer a promising unsupervised interpretability approach for extracting human-interpretable concepts from large language models. Yet, their use in the security domain remains underexplored. Security-related classification tasks typically rely on smaller models than those commonly studied with SAEs. In this paper, we examine how SAEs can be used to interpret classification models fine-tuned for security tasks. We apply an interpretability framework that combines established techniques for foundation models to generate concept explanations, focusing on two widely studied problems in safety and security: hate speech and deepfake detection. We demonstrate its ability to produce meaningful concept explanations while identifying critical challenges for the effective deployment of SAEs in security contexts. Our findings suggest that while SAEs offer a promising unsupervised technique for generating concept explanations, addressing the identified challenges is necessary for their useful application in security interpretability.
Nils Ole Breuer, L. Linhardt, Philipp Normann et al.· 2026 IEEE European Symposium...· 0 citations
A new perspective of information theory is introduced to investigate LVLMs’ transferable characteristics by exploring the relative dependence between outputs of the LVLM model and input adversarial samples and formulate the complicated calculation of information gain as an estimation problem and incorporate such informative constraints into the adversarial learning process.
Xiaowen Cai, Daizong Liu, Xiaoye Qu et al.· Neural Information Processin...· 7 citations
This paper proposes a novel LVLM attack method, called BadPhase with further backdoor designs, to implant adversarial phase as triggers into any image inputs via data poisoning so as to control the LVLMs’ predictions and finds that LVLMs are sensitive to the phase-aware image structure.
Daizong Liu, Junhao Dong, Xiang Fang et al.· 0 citations
Cloud-based language model services routinely process prompts containing sensitive information. Obfuscation-based defenses---including ObfusLM, SentinelLMs, TextObfuscator, and DPNR---mitigate this risk by transforming prompt representations before transmission, offering a lightweight alternative to cryptographic solutions. We show these defenses provide far less protection than previously believed. We present DeepInvert, a semi-supervised embedding inversion attack that recovers original tokens from obfuscated representations with higher accuracy than prior methods. The key insight is that unlabeled obfuscated embeddings retain exploitable semantic structure despite perturbation. DeepInvert combines supervised training on labeled shadow data with a novel unsupervised consistency objective over unlabeled target embeddings, alternating between the two via a mixed training pipeline. Defense-aware adaptations further extend the attack to diverse obfuscation mechanisms across encoder-based and autoregressive architectures. Experiments on nine defenses, five tasks, and four model architectures show that DeepInvert outperforms prior attacks on most defenses. Against ObfusLM, DeepInvert achieves 73.5\% top-1 token recovery versus 26.2\% for the previous best. Our results reveal a task-dependent tension: obfuscation schemes preserving enough signal for utility also retain sufficient structure for inversion, while schemes resisting inversion collapse utility. On simpler classification tasks, some DP-based defenses can maintain both. We call for a re-evaluation of this defense class.
The Explainable Deepfake Detection Challenge at ACM Multimedia 2026 is designed to benchmark this joint capability of classification metrics with semantic similarity, simplicity, and intent-aware grounding metrics that assess whether explanations identify the relevant manipulated entities and supporting visual evidence.
Abhijeet Narang, Kartik Kuckreja, Shreya Ghosh et al.· 1 citation
Generative Adversarial Networks or GANs, have become a significant approach in deep learning along with Con-volutional and Recurrent Neural Networks, due to improvements in computing technology and more advanced ways to train these frameworks or models. Since GANs were first introduced in 2014, their application has expanded beyond image generation to include critical security tasks like anomaly detection and malware analysis. This paper offers a comprehensive survey of how GAN-based methods are utilized for identifying unusual and harmful activities in cyber settings. It examines key variants of GANs relevant to this field, explains their fundamental architectures and training methods, and explains their integration into systems to detect anomalies and malware. Additionally, the paper catalogs publicly accessible datasets and evaluation metrics frequently used in the reviewed studies to illustrate common experimental methodologies and research directions. Finally, it addresses ongoing challenges and potential future avenues for employing GANs to counteract emerging cybersecurity threats, highlighting their importance in developing more proactive and robust security measures.
A. Thakore, Neha Gupta, Akash Saxena et al.· International Journal of Inn...· 0 citations