New benchmark reveals flaws in editing medical AI models on the fly

Original: Evaluating and Understanding Model Editing for Medical Vision Language Models

Source: arxiv.org ↗