Công bố khoa học

PoCForge: A Closed-Loop Multi-Agent Framework for Autonomous Exploit Synthesis and Verification

Penetration testing remains a critical yet resource-intensive task in modern cybersecurity practice, as traditional approaches heavily rely on expert-driven manual workflows that are difficult…

Improving TTP Alignment in LLM-Based Malware Artifact Generation via Hybrid Verification

DriftGuard-APT: Concept Drift-Aware Unsupervised Graph Learning for APT Detection on Provenance Streams

Action-Graph: A Graph-Aware Multi-Step LLM Pipeline for MITRE ATT&CK TTP Extraction from Cyber Threat Intelligence

Evidence-Preserving Multi-Agent LLMs for SOC Threat Hunting with Different Collaboration and Prompting Strategies

Interpretable Detection and Threat Characterization of Malicious PowerShell Scripts via Multi-Level Representation Fusion and LLM Reasoning

Malicious PowerShell scripts remain a practical challenge for enterprise security because attackers can combine obfuscation, fileless execution, and legitimate administrative utilities to evade detectors…

ContractShield: Bridging Semantic-Structural Gaps via Hierarchical Cross-Modal Fusion for Multi-Label Vulnerability Detection in Obfuscated Smart Contracts

Smart contracts are increasingly targeted by adversaries employing obfuscation techniques such as bogus code injection and control-flow manipulation to evade vulnerability detection. Existing multimodal…

FedAug-CTI: A Federated Framework with LM-Based Cyber Threat Intelligence Generation for Hierarchical Tactic-Technique Mapping

Mapping unstructured Cyber Threat Intelligence (CTI) reports to MITRE ATT\&CK Tactics and Techniques is essential for threat understanding but remains largely manual due to…

SSCFuzz: Combining LLM-Generated Transaction Sequences with Undirected Exploration for Smart Contract Fuzzing

Smart contracts manage substantial financial value yet resist testing: valid inputs follow an application binary interface (ABI), exploitable bugs need multi-step stateful sequences, and…