Counterfactual Modeling with Fine-Tuned LLMs for Health Intervention Design and Sensor Data Augmentation

Abstract

Counterfactual explanations (CFEs) provide human-centric interpretability by identifying the minimal, actionable changes required to alter a machine learning model’s prediction. Therefore, CFs can be used as (i) interventions for abnormality prevention and (ii) augmented data for training robust models. we conduct a comprehensive evaluation of CF generation using large language models (LLMs), including GPT-4 (zero-shot and few-shot) and two open-source models-BioMistral-7B and LLaMA-3.1-8B-in both pretrained and fine-tuned configurations. Using the multimodal AI-READI clinical dataset, we assess CFs across three dimensions: intervention quality, feature diversity, and augmentation effectiveness. Fine-tuned LLMs, particularly LLaMA-3.1-8B, produce CFs with high plausibility (up to 99%), strong validity (up to 0.99), and realistic, behaviorally modifiable feature adjustments. When used for data augmentation under controlled label-scarcity settings, LLM-generated CFs substantially restore classifier performance, yielding an average 20% F1 recovery across three scarcity scenarios. Compared with optimization-based baselines such as DiCE, CFNOW, and NICE, LLMs offer a flexible, model-agnostic approach that generates more clinically plausible, behaviorally actionable and semantically coherent counterfactuals. Overall, this work demonstrates the promise of LLM-driven counterfactuals for both interpretable intervention design and data-efficient model training in sensor-based digital health.

Shovito Barua Soumma

Graduate Research Associate

I am a Ph.D. candidate at Arizona State University. I work as a Graduate Research Associate at Embedded Machine Intelligence Lab (EMIL) under the supervision of Dr. Hassan Ghasemzadeh.

Asiful Arefeen

Graduate Research Assistant

I am a PhD student at Arizona State University (ASU). I am working under the supervision of Professor Hassan Ghasemzadeh at the Embedded Machine Intelligence Lab (EMIL). I am interested in explainable AI, using AI to generate interventions in digital health, machine learning, passive sensing and mobile health. I received a BS in Electrical & Electronic Engineering from Bangladesh University of Engineering & Technology (BUET) in 2019 and an MS in Biomedical Informatics from Arizona State University in 2023.

Hassan Ghasemzadeh

Director

Hassan Ghasemzadeh is an Associate Professor of Biomedical Informatics at Arizona State University (ASU) and a Computer Science Adjunct Faculty at Washington State University (WSU).