ECG-Reasoning-Benchmark: A Benchmark for Evaluating Clinical Reasoning Capabilities in ECG Interpretation

arXiv:2603.14326 · cs.LG, cs.AI, cs.CL · Submitted 2026-03-15 · Read on arXiv

cs.LG, cs.AI, cs.CL

Submitted: 2026-03-15

Updated: 2026-10-28

Comments: Accepted at NeurIPS 2026 Evaluations & Datasets Track

Code: https://github.com/Jwoo5/ecg-reasoning-benchmark

License: http://creativecommons.org/licenses/by/4.0/

The gist: While Multimodal Large Language Models (MLLMs) show promising performance in automated electrocardiogram interpretation…

Terminology

Abstract

While Multimodal Large Language Models (MLLMs) show promising performance in automated electrocardiogram interpretation, it remains unclear whether they perform actual step-by-step reasoning or just rely on superficial visual cues. To investigate this, we introduce ECG-Reasoning-Benchmark, a novel multi-turn evaluation framework comprising over 6,400 samples to systematically assess step- by-step reasoning across 17 core ECG diagnoses. Our comprehensive evaluation of state-of-the-art models reveals a systematic shortfall in executing multi-step logical deduction. Although models possess the medical knowledge to retrieve clinical criteria for a diagnosis, they exhibit consistently low success rates in maintaining a complete reasoning chain, primarily failing to ground the corresponding ECG findings to the actual visual evidence in the ECG signal. These results demonstrate that current MLLMs bypass actual visual interpretation, highlighting a limitation in their visual grounding capabilities and motivating robust, reasoning-centric medical AI. The code and data are available at https://github.com/Jwoo5/ecg-reasoning-benchmark.

Sources

Related papers