Virginia Tech · Computer Science
Trustworthy AI. Resilient systems. Safer communities.
Welcome to tClab
We study how AI, cybersecurity, and human-centered design can make digital and autonomous systems more trustworthy, resilient, and safe.
Directed by Dr. Jin-Hee Cho, tClab connects trustworthy and adversarial machine learning, decision-making under uncertainty, and network science to real-world challenges in critical infrastructure and online safety.
Our researchMeet the labCongrats Heajun and the team! “Grounded but Misleading: Evaluating Semantic Alignment in AI-Generated Security Explanations” will be presented at the 9th AAAI Conference on AI, Ethics, and Society (AIES), Oct. 12–14, 2026.
Congrats K. Li and the team! “Evidential Semantic Uncertainty Decomposition for Large Language Models” accepted to NeurIPS 2026.
Dr. Cho’s CogSec'2026 keynote: “AI, Influence, and Collective Sensemaking: Beyond Binary Beliefs: Uncertainty-Aware Opinion Dynamics for Understanding, Resisting, and Countering Disinformation.”
Dr. Cho’s CLUSTER'2026 talk: “The Price of Privacy: Fairness–Utility Trade-offs in Responsible Federated Learning.”
Congrats Heajun and the team! “StagePilot: Stage-Level Planning for Long-Horizon Dialogue Simulation in Cybergrooming” was presented at SIGDIAL 2026;
Congrats Dawood and the team! “SHIELD: Secure Human-Machine Interaction with Evidential Learning and Dynamic Trust for Drone Swarm Control” was presented at USENIX VehicleSec 2026;
Congrats Dawood and the team! “RAIL: Risk-Aware Human-in-the-Loop Framework with Adaptive Intrusion Response for Autonomous Vehicles” was presented at USENIX VehicleSec 2026. Google Scholar
Congrats Dawood and the team! “Explainable Federated Learning via Global–Local Attribution Alignment” was presented at ICML 2026 in Korea. Google Scholar
Privacy-preserving, fair, explainable, and uncertainty-aware AI for smart farms and healthcare. Research combines secure wireless sensing, federated learning, and failure detection to support reliable decisions in critical systems.
Explore project website →AI-driven mission assurance, adaptive intrusion response, and cyberdeception for autonomous vehicles and tactical networks. Uncertainty-aware reasoning and human-machine teaming connect security decisions to mission performance.
Explore project website →Conversational AI and experiential learning to help adolescents recognize cybergrooming risks, strengthen self-efficacy, and respond to risky online situations. This interdisciplinary effort connects AI, cybersecurity, and youth-centered safety education.
Explore project website →Research on AI reliability, behavioral cybersecurity data, evidence-grounded explanations, and human-centered cyber safety. Projects examine bias and uncertainty and develop approaches to scam prevention and training for vulnerable populations, including older adults.
Explore project website →Browse papers and citation information on Google Scholar.
View Google ScholarInterested in research on trustworthy AI and cybersecurity? Learn about our research and get in touch to discuss potential opportunities.
Joining the lab →