AC - Research

Visual Analytics for Explainable and Trustworthy AI

I develop visual analytics systems that help AI experts understand, inspect, and evaluate AI models, such as tabular foundation models for structured data. My research explores model behavior, explanation reliability, and concept-based explainability through interactive approaches that enable domain experts to create, refine, and compare concepts, as well as reason about and with complex AI models.

Dimensionality Reduction and High-Dimensional Data Analysis

I develop visual and interactive approaches for understanding high-dimensional data, with a particular focus on dimensionality reduction (DR). My research investigates how to assess, interpret, and explain projections and clusters, helping analysts identify meaningful structure while understanding the distortions, uncertainty, and limitations introduced by DR techniques.

Human-Centered AI for Decision Making

I investigate how AI and visualization can support people facing complex, uncertain, and ill-defined decisions. My research emphasizes human agency and deliberation, exploring how interactive systems can help people express evolving preferences, reason about uncertainty, examine alternatives, and progressively align human judgment with computational representations.

Human-AI Interaction with Generative Models

I study how people understand, evaluate, and interact with generative AI models and their outputs. My research explores LLM-assisted interpretation of complex data, interactive explanation and control, and human assessment of AI-generated media, with the goal of making generative systems more understandable, reliable, and useful in human-centered analytical tasks.

Connect at

Page updated

Google Sites

Report abuse