Công bố khoa học

VulForge: An Automatic Script Generation Framework for Vulnerability Deployment in Honeypots Using Large Language Models

Honeypots play a critical role in cyber defense by emulating systems that attract and deceive attackers, allowing defenders to monitor malicious behavior in a…

FraudTrace: Verifying Fraudulent News to Prevent Online Scam Campaigns via a Multi-Agent LLM-Based System

Multimodal Learning with LLM-Fusion Head for Android Malware Detection: Enhancing Cross-Modality Robustness Against Obfuscation and Adversarial Samples

X-AdvIDS: A framework for assessing and improving the adversarial robustness of Intrusion Detection Systems with explainability-guided mutation and analysis

UniFuzz: A Unified Fuzzing Framework for Smart Contract Vulnerability Detection with LLMs and Audit Reports

Multimodal Fusion of Behavior Graphs and API Sequences for Ransomware Detection

Multimodal Windows Malware Detection via Hybrid Analysis and Enriched Graphs: Effectiveness and Explainability

Multimodal Windows Malware Detection via Hybrid Analysis and Enriched Graphs: Effectiveness and Explainability

Smart Contract Vulnerability Detection using Prompt Engineering with Reasoning Models

The increasing number of smart contracts has gained significant attention to the urgency of robust and scalable vulnerability detection techniques to mitigate substantial financial…

PowerGNN: A source code structural and textual awareness approach for identifying malicious PowerShell scripts