sebis at ArchEHR-QA 2026: How Much Can You Do Locally? Evaluating Grounded EHR QA on a Single Notebook

TL;DR

This paper demonstrates that privacy-preserving, local EHR question answering systems can achieve competitive performance using commodity hardware, enabling secure clinical data access without cloud reliance.

Contribution

It evaluates the feasibility of grounded EHR QA on a single notebook, showing that smaller models can perform well when properly configured, and participates in all shared task subtasks.

Findings

Local systems can achieve competitive performance on shared tasks.

Smaller models can approach larger system performance with proper configuration.

Fully local EHR QA systems are feasible with current models and hardware.

Abstract

Clinical question answering over electronic health records (EHRs) can help clinicians and patients access relevant medical information more efficiently. However, many recent approaches rely on large cloud-based models, which are difficult to deploy in clinical environments due to privacy constraints and computational requirements. In this work, we investigate how far grounded EHR question answering can be pushed when restricted to a single notebook. We participate in all four subtasks of the ArchEHR-QA 2026 shared task and evaluate several approaches designed to run on commodity hardware. All experiments are conducted locally without external APIs or cloud infrastructure. Our results show that such systems can achieve competitive performance on the shared task leaderboards. In particular, our submissions perform above average in two subtasks, and we observe that smaller models can approach the performance of much larger systems when properly configured. These findings suggest that privacy-preserving EHR QA systems running fully locally are feasible with current models and commodity hardware. The source code is available at https://github.com/ibrahimey/ArchEHR-QA-2026.