Paper: Reasoning About Reasoning: Towards Informed and Reflective Use of LLM Reasoning in HCI
A worksheet with a set of structured questions to facilitate critical reflection on using LLMs' "reasoning-like" behavior in HCI systems.
Paper: Argument-Based Consistency in Toxicity Explanations of LLMs (EACL '26)
A suite of metrics and codebase to evaluate the logical consistency of free-form reasoning provided by Large Language Models on toxicity tasks.
Paper: Talking About the Assumption in the Room (CHI '25)
A structured framework to support Machine Learning practitioners and researchers in articulating implicit assumptions, as premises to arguments, in their workflows.
Paper: Towards a Non-Ideal Methodological Framework for Responsible ML (CHI '24)
A framework and visualizer tool for diagnosing non-ideal conditions in machine learning workflows and mapping them to abstract values and technical interventions.
Paper: Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations (FAccT '20)
Codebase to generate a set of counterfactual explanations for any Machine Learning model.