Research & Academic Work

My research asks a question generative AI has made urgent: when can a simulated human decision-maker be trusted? I approach it by owning every link of the evidence chain. I designed and fielded a 937-household flood-adaptation survey in New Jersey’s Passaic River Basin, used structural equation modeling to identify how owners and renters actually decide among adaptation actions, and encoded those empirically identified pathways in a Bayesian-calibrated agent-based model of 52,141 households coupled to a catastrophe flood model with National Flood Insurance Program mechanics — including income-normalized equity analysis of who bears flood losses. With my advisor, I published a Water Resources Research article (62(6), e2025WR042111, 2026) substituting large language models as the behavioral engine of such simulations. That substitution is the contribution: large language models become the behavioral engine of the simulation. Because nothing guarantees a generative agent decides the way real households do, my current work pairs it with AI governance — checking LLM personas against the decision pathways identified in my own survey data, and developing governance methods that constrain agent behavior with physical constraints and behavioral theory — research software now in preparation for release. First-author manuscripts on the coupled model and on adaptation decision pathways are under review. The longer-term agenda is a governance standard for generative agents in consequential simulation.

The research program, as one arc

  1. Survey

    Designed and fielded a 937-household flood-adaptation survey (557 owners / 379 renters) in New Jersey’s Passaic River Basin.

    Evidence Calibration files in the FLOODABM repository; the empirical grounding of the Water Resources Research paper.

    Instrument feeding one published article and manuscripts now under review.

  2. Decision pathways (SEM)

    Identified flood-adaptation decision pathways through structural equation modeling — across owner and renter groups, and across marginalized and non-marginalized groups (the equity strand).

    Evidence First-author manuscript targeting Sustainable Cities and Society.

    Under review.

  3. Coupled ABM–CAT model

    Built a Bayesian-calibrated agent-based model of 52,141 households across 27 census tracts, coupled to a catastrophe flood model with National Flood Insurance Program premium, payout, and deductible mechanics and income-normalized equity analysis of who bears flood losses.

    Evidence FLOODABM repository — Zenodo-archived with CITATION.cff, published seed lists, and a candid known-limitations register; first-author manuscript targeting the Journal of Hydrology.

    Code: Research prototype, archived. Paper: Under review.

  4. LLMs as the behavioral engine

    Applied large language models as the behavioral engine of an agent-based model of a human–water system — the core contribution. Because nothing guarantees a generative agent decides the way real households do, the work also checks LLM personas against empirically identified human decision pathways.

    Evidence Yang, Y. C. E., & Chiou, W. (2026). Leveraging Large Language Models for Agent-Based Simulation of Human–Water System Interactions. Water Resources Research, 62(6), e2025WR042111.

    Published.

  5. Governance methods

    Validation architecture for generative agents — hard physical and financial constraints plus behavioral-theory coherence, fail-closed, with audit trails — generalized beyond the flood domain.

    Evidence Research software in preparation for release; public echoes in the fail-closed design of the open-source tooling.

    In preparation.

Publications & talks

Wenyu Chiou standing beside his AGU Fall Meeting 2025 poster, NH41E-0449.
Presenting poster NH41E-0449 — Modeling Long-Term Household Flood Adaptation under Social Heterogeneity — at the AGU Fall Meeting 2025.