A selection of representative work from the lab. Several papers are currently under review; the full list is updated with links and DOIs as papers are published.
I. Lee
Dynamic DNN partitioning and cooperative offloading across heterogeneous devices on EV-Hub edge infrastructure, optimizing latency and energy for public-safety networks.
I. Lee
A Victim–Attacker–EchoChamber generative agent-based model that pre-tests voice-phishing defense strategies for older adults under adversarial coevolution.
I. Lee
Distributional fidelity, stereotype audit, convergent and construct validity — four pillars that tier synthetic personas for reliable research use.
I. Lee
Quantifying self-preference bias in LLM-as-judge evaluation across procedural scenarios and alignment conditions.
I. Lee
Agent-based simulation comparing counter-disinformation interventions, revealing emergent backfire effects and adversarial coevolution dynamics.
I. Lee
An impossible-trinity lens on policing AI, developed through a synthetic expert panel and 2×2 scenario planning across seven national cases.