AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

MedFG-VQA: Low-Frequency Memory and Graph Attention for Lightweight Medical VQA

Haowen Gu, Gensheng Pei, Zeren Sun, Mingwu Ren, Xiangbo Shu, Yazhou Yao, Fumin Shen

CVPR 2026 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add FMF and GACA modules (ablation)MedFG-VQA baseline · SynMedVQA averageaccuracy: 0.627 -> 0.6441Improved
use DCT instead of FFT in FMF (ablation)FMF module · SynMedVQA averageaccuracy: 0.5585 -> 0.6441Improved
add DCT frequency transform versus none (ablation)FMF module · SynMedVQA averageaccuracy: 0.6058 -> 0.6441Improved
use FFT instead of no transform (ablation)FMF module · SynMedVQA averageaccuracy: 0.6058 -> 0.5585Hurt
replace standard cross-attention with GACA (ablation)cross-modal attention · SynMedVQA averageaccuracy: 0.6407 -> 0.6441Improved
increase memory bank size from 16 to 64 (ablation)FMF memory · SynMedVQA averageaccuracyImproved
increase memory bank size from 64 to 128 (ablation)FMF memory · SynMedVQA averageaccuracyHurt
vary diversity loss weight lambda from 0.3 to 0.9 (ablation)FMF diversity loss · SynMedVQA averageaccuracy: 0.6441No clear effect
MedFG-VQA vs Qwen3-VL on SynMedVQAQwen3-VL · SynMedVQA averageaccuracy: 0.5492 -> 0.6441Improved
MedFG-VQA vs Qwen3-VL on closed-ended VQA-RADQwen3-VL · VQA-RAD closedaccuracy: 0.7868 -> 0.6324Hurt
MedFG-VQA vs Qwen3-VL on open-ended PathVQAQwen3-VL · PathVQA openaccuracy: 0.5249 -> 0.8062Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy