DATE: Friday, October 16, 2026
TIME: 11:45 AM to 1:00 PM (ET)
Location: Discovery House, 6th Floor Conference Room
Virtual Attendance: Available via Zoom, registration required – Register Here
Lunch: Served at 11:45 AM
The Build America Center (BAC) and the UMD-Future Cities Cohort invite researchers, students, and professionals to an upcoming seminar titled “Context Modeling: Large Language Models as Research Methods,” featuring Dr. Abdolmajid (Mazi) Erfani, Assistant Professor of Civil & Environmental Engineering at the University of Washington.
The seminar will take place on Friday, October 16, 2026, from 11:45 AM to 1:00 PM (ET) at Discovery House, 6th Floor Conference Room, with a virtual attendance option available via Zoom. Lunch will be served beginning at 11:45 AM.
The presentation will explore how large language models (LLMs) can be used to study human behavior and decision-making through an emerging research approach known as Context Modeling. This structured method translates real-world data and decision-making environments into contextual representations for LLM-based simulations.
Dr. Erfani will discuss how Context Modeling integrates research design, persona construction, controlled prompting, and systematic validation to evaluate the fidelity, robustness, consistency, and reproducibility of LLM-based experiments. Drawing on applications in infrastructure decision-making, building occupant behavior, and travel simulation, the session will demonstrate how researchers can design and assess LLM-based experiments, interpret their findings, and recognize the limitations of these emerging methods.
About the Speaker
Dr. Abdolmajid (Mazi) Erfani is an Assistant Professor of Civil & Environmental Engineering at the University of Washington and Director of the Analytics for Infrastructure Management (AIM) Lab. His research focuses on applying artificial intelligence and data analytics to improve decision-making in construction, transportation, and infrastructure systems. He earned his Ph.D. in Civil Engineering from the University of Maryland, College Park, in 2023. His work aims to advance intelligent, resilient, and human-centered infrastructure planning and management.
Researchers, students, and professionals interested in artificial intelligence, infrastructure systems, and innovative research methodologies are encouraged to attend this seminar to learn more about the potential of LLMs as tools for studying complex human decision-making processes.
