Awenbocc/GEMeX-Project

Official code of paper "GEMeX: A Large-Scale, Groundable, and Explainable Medical VQA Benchmark for Chest X-ray Diagnosis" [ICCV 2025]

Jupyter Notebook

25 commits

updated Jun 29, 2025

www.med-vqa.com/GEMeX

See the code

README

GEMeX: A Large-Scale, Groundable, and Explainable Medical VQA Benchmark for Chest X-ray Diagnosis

GEMeX Display

News

[2025/06/28] 🧠 Building on GEMeX, we now introduce ThinkVG, which helps LVLMs think while perform visually grounding.
[2025/06/26] 🎉 Great News! Our paper has been accepted to ICCV 2025!
[2025/01/17] We release pre-trained LLaVA-Med-GEMeX, a simple baseline model based on GEMeX VQA data.
[2024/12/01] We release part of data and complete training code.

How to fine-tune:

git clone https://github.com/Awenbocc/GEMeX-Project.git
cd llava-med
bash fine_tune_gemex.sh # modify data location

medical-vision-language-model

Similar repos

BoKelvin/GEMeX-VQA-Model-Simple

Model

LLaVA-Med-GEMeX (v1) fine-tuned with LLaVA-Med on the GEMeX-VQA dataset (www.med-vqa.com/GEMeX)

Cocofeat/uMedGround

【IEEE TPAMI 2025】Uncertainty-aware Medical Diagnostic Phrase Identification and Grounding

Keda710/Explanability_in_VLM

MSIIP/MedM-VL

MedM-VL is a modular, LLaVA-based codebase for medical LVLMs.

uni-medical/GMAI-VL

GMAI-VL & GMAI-VL-5.5M: A Large Vision-Language Model and A Comprehensive Multimodal Dataset…

102

function2-llx/MMMM

[NAACL 2025] VividMed: Vision Language Model with Versatile Visual Grounding for Medicine

icarus402/llava-dino

XYPB/GLAM

Official Repo for MICCAI 25 Paper: Geometry-Guided Local Alignment for Multi-View Visual Language…