I'm a Computer Science PhD student at Virginia Tech, advised by Dr. Jiaming Cui and affiliated with the Sanghani Center. I'm also a PRISM AI Safety Research Fellow.
MY RESEARCH · IN BRIEF
I study multi-agent safety and coordination, self-evolving agents, and emergent behaviors in LLM-based multi-agent systems. My work investigates how agents communicate, delegate, and develop collective patterns; how failures propagate through memory, tools, and communication; and how adaptive agent harnesses can help us monitor, evaluate, and safeguard increasingly autonomous systems.
More broadly, I'm interested in the relationship between agent autonomy, collective behavior, and system-level reliability—both empirically and through methods that make these systems more interpretable and dependable.
I have also worked on generative vision and diffusion models, exploring rare-concept image generation, personalized and creative editing, and ways of combining language and visual reasoning.
Before starting my PhD, I earned an M.S. in Data Science from Northeastern University in Boston and spent roughly two and a half years working as a data scientist.
Outside the lab, I'm also a writer and editor. Poetry, philosophy, photography, and bringing people together to learn are all part of this little world.
My research explores how language-model agents interact, evolve, and develop collective behaviors—and the mechanisms needed to make increasingly autonomous agentic systems safe and reliable.
I study how language-model agents communicate, delegate work, establish dependencies, and manage trust across distributed workflows. My broader interests include secure coordination, cascading failures, adversarial interaction, and the reliability of collaborative systems under uncertainty.
The four research directions are visible on the wheel; the moving marker changes the illustration and the note automatically. Select a direction to hold it for a moment.
A few pieces of the research story, illustrated with short loops from the papers themselves.
Online Detection and Attribution of Cascade Attacks in LLM Multi-Agent Systems via Cross-Channel Causal Monitoring
Online detection and causal attribution of cascading attacks across multi-agent communication, memory, and tool channels.
The Hidden Power Laws of Collective Cognition in LLM Multi-Agent Systems
Empirical power-law patterns in collective cognition, influence concentration, and coordination across more than 1.5 million agent interactions.
Physics-Guided Generative Agent-Based Modeling
Mechanism-guided generative agent-based simulation with hierarchical populations, aimed at scalable and more calibrated collective modeling.
Rare Concept Generation and Editing via Graph-driven Relational Guidance
Training-free rare-concept image generation and editing using graph-guided relational context and iterative correction.
Further papers and full publication details are in my bibliography.
Browse all publicationsNew notes pinned to the board, and an old-news drawer for everything that brought me here.
Honored to receive the Outstanding Reviewer recognition from CVPR 2026.
Agentic Web, SocialAgent, AI4SCk, and AI4Good; grateful to the collaborators behind each project.
Two very different research directions—multi-agent safety and generative visual modeling—heading to the main conference.
Completed my Machine Learning Research internship in Seattle.
Working on decision reliability under conflicting, changing, and uncertain evidence.
New research on collective cognition, influence, and coordination at scale.
Exploring scalable generative agent-based simulation and structured collective modeling.
Celebrating two accepted papers at NeurIPS 2025.
Completed a summer research internship at Amazon Science.
Showing 9 notes
Released the CREA preprint. CREA preprint ↗
Presented research at CAPWIC 2025 in Washington, DC. CAPWIC 2025 ↗
FluxSpace accepted to CVPR 2025. FluxSpace ↗
Released Context Canvas and FluxSpace on arXiv. Context Canvas ↗
Started my PhD in Computer Science at Virginia Tech.
Hosted a two-day Generative AI workshop at Northeastern University as a WiDS Ambassador. Northeastern ↗
Invited talk on responsible AI at UMass Amherst. UMass Amherst ↗
Appointed a WiDS Worldwide Ambassador. WiDS ↗
Hosted an Analytics Vidhya DataHour webinar on LLMs. DataHour ↗
Invited talk at Northeastern University on MLOps.
Joined Fidelity Investments as a Data Scientist. Fidelity ↗
Graduated with an M.S. in Data Science from Northeastern University, Boston.
FROM THE PERSONAL ARCHIVE