AMANDA: Agentic Medical Knowledge Augmentation for Data-Efficient Medical Visual Question Answering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Ziqing, Mao, Chengsheng, Wen, Xiaole, Luo, Yuan, Ding, Kaize
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911189237235712
author Wang, Ziqing
Mao, Chengsheng
Wen, Xiaole
Luo, Yuan
Ding, Kaize
author_facet Wang, Ziqing
Mao, Chengsheng
Wen, Xiaole
Luo, Yuan
Ding, Kaize
contents Medical Multimodal Large Language Models (Med-MLLMs) have shown great promise in medical visual question answering (Med-VQA). However, when deployed in low-resource settings where abundant labeled data are unavailable, existing Med-MLLMs commonly fail due to their medical reasoning capability bottlenecks: (i) the intrinsic reasoning bottleneck that ignores the details from the medical image; (ii) the extrinsic reasoning bottleneck that fails to incorporate specialized medical knowledge. To address those limitations, we propose AMANDA, a training-free agentic framework that performs medical knowledge augmentation via LLM agents. Specifically, our intrinsic medical knowledge augmentation focuses on coarse-to-fine question decomposition for comprehensive diagnosis, while extrinsic medical knowledge augmentation grounds the reasoning process via biomedical knowledge graph retrieval. Extensive experiments across eight Med-VQA benchmarks demonstrate substantial improvements in both zero-shot and few-shot Med-VQA settings. The code is available at https://github.com/REAL-Lab-NU/AMANDA.
format Preprint
id arxiv_https___arxiv_org_abs_2510_02328
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AMANDA: Agentic Medical Knowledge Augmentation for Data-Efficient Medical Visual Question Answering
Wang, Ziqing
Mao, Chengsheng
Wen, Xiaole
Luo, Yuan
Ding, Kaize
Computation and Language
Artificial Intelligence
Multiagent Systems
Medical Multimodal Large Language Models (Med-MLLMs) have shown great promise in medical visual question answering (Med-VQA). However, when deployed in low-resource settings where abundant labeled data are unavailable, existing Med-MLLMs commonly fail due to their medical reasoning capability bottlenecks: (i) the intrinsic reasoning bottleneck that ignores the details from the medical image; (ii) the extrinsic reasoning bottleneck that fails to incorporate specialized medical knowledge. To address those limitations, we propose AMANDA, a training-free agentic framework that performs medical knowledge augmentation via LLM agents. Specifically, our intrinsic medical knowledge augmentation focuses on coarse-to-fine question decomposition for comprehensive diagnosis, while extrinsic medical knowledge augmentation grounds the reasoning process via biomedical knowledge graph retrieval. Extensive experiments across eight Med-VQA benchmarks demonstrate substantial improvements in both zero-shot and few-shot Med-VQA settings. The code is available at https://github.com/REAL-Lab-NU/AMANDA.
title AMANDA: Agentic Medical Knowledge Augmentation for Data-Efficient Medical Visual Question Answering
topic Computation and Language
Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2510.02328