Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yang, Jiayan, Wu, Zhuoyu, Fang, Wenqi
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.15561
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911687699857408
author Yang, Jiayan
Wu, Zhuoyu
Fang, Wenqi
author_facet Yang, Jiayan
Wu, Zhuoyu
Fang, Wenqi
contents Vision-Language Models (VLMs) facilitate medical visual question answering (MedVQA) by jointly interpreting images and text. However, existing models typically depend on large architectures and closed-set answers, which limits their efficiency and potential clinical applicability. To overcome these shortcomings, we introduce RoiMAM, an efficient VLM. It integrates a training-free ROI Generation Module with Semantic Selective Suppression to focus on lesion-relevant regions, alongside a Text Prompt Enhancer module that provides modality-specific context without introducing training parameters. Compared to the widely used MedVInT-TD model, our design achieves efficient and accurate diagnosis at less than 20\% of the model size, while improving accuracy by approximately 2% on SLAKE and 4.6% on PMC-VQA.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15561
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RoiMAM: Region-of-Interest Medical Attention Model for Efficient Vision-Language Understanding
Yang, Jiayan
Wu, Zhuoyu
Fang, Wenqi
Computer Vision and Pattern Recognition
Vision-Language Models (VLMs) facilitate medical visual question answering (MedVQA) by jointly interpreting images and text. However, existing models typically depend on large architectures and closed-set answers, which limits their efficiency and potential clinical applicability. To overcome these shortcomings, we introduce RoiMAM, an efficient VLM. It integrates a training-free ROI Generation Module with Semantic Selective Suppression to focus on lesion-relevant regions, alongside a Text Prompt Enhancer module that provides modality-specific context without introducing training parameters. Compared to the widely used MedVInT-TD model, our design achieves efficient and accurate diagnosis at less than 20\% of the model size, while improving accuracy by approximately 2% on SLAKE and 4.6% on PMC-VQA.
title RoiMAM: Region-of-Interest Medical Attention Model for Efficient Vision-Language Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.15561