MultiMedEdit: A Scenario-Aware Benchmark for Evaluating Knowledge Editing in Medical VQA

TL;DR

MultiMedEdit introduces a comprehensive benchmark for evaluating knowledge editing in multimodal medical visual question answering, addressing the unique challenges of integrating updated knowledge with visual reasoning in clinical scenarios.

Contribution

It is the first benchmark specifically designed for knowledge editing in clinical multimodal tasks, including a new metric suite and extensive experimental analysis.

Findings

Current methods struggle with generalization and long-tail reasoning.

Significant practical trade-offs exist in edit latency and memory footprint.

The benchmark reveals limitations of existing approaches in complex clinical workflows.

Abstract

Knowledge editing (KE) provides a scalable approach for updating factual knowledge in large language models without full retraining. While previous studies have demonstrated effectiveness in general domains and medical QA tasks, little attention has been paid to KE in multimodal medical scenarios. Unlike text-only settings, medical KE demands integrating updated knowledge with visual reasoning to support safe and interpretable clinical decisions. To address this gap, we propose MultiMedEdit, the first benchmark tailored to evaluating KE in clinical multimodal tasks. Our framework spans both understanding and reasoning task types, defines a three-dimensional metric suite (reliability, generality, and locality), and supports cross-paradigm comparisons across general and domain-specific models. We conduct extensive experiments under single-editing and lifelong-editing settings. Results suggest that current methods struggle with generalization and long-tail reasoning, particularly in complex clinical workflows. We further present an efficiency analysis (e.g., edit latency, memory footprint), revealing practical trade-offs in real-world deployment across KE paradigms. Overall, MultiMedEdit not only reveals the limitations of current approaches but also provides a solid foundation for developing clinically robust knowledge editing techniques in the future.