Saved in:
Bibliographic Details
Main Authors: Chen, Zhihui, He, Kai, Lei, Qingyuan, Pu, Bin, Zhang, Jian, Xu, Yuling, Feng, Mengling
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.18577
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908900735844352
author Chen, Zhihui
He, Kai
Lei, Qingyuan
Pu, Bin
Zhang, Jian
Xu, Yuling
Feng, Mengling
author_facet Chen, Zhihui
He, Kai
Lei, Qingyuan
Pu, Bin
Zhang, Jian
Xu, Yuling
Feng, Mengling
contents Text-guided image editors can now manipulate authentic medical scans with high fidelity, enabling lesion implantation/removal that threatens clinical trust and safety. Existing defenses are inadequate for healthcare. Medical detectors are largely black-box, while MLLM-based explainers are typically post-hoc, lack medical expertise, and may hallucinate evidence on ambiguous cases. We present MedForge, a data-and-method solution for pre-hoc, evidence-grounded medical forgery detection. We introduce MedForge-90K, a large-scale benchmark of realistic lesion edits across 19 pathologies with expert-guided reasoning supervision via doctor inspection guidelines and gold edit locations. Building on it, MedForge-Reasoner performs localize-then-analyze reasoning, predicting suspicious regions before producing a verdict, and is further aligned with Forgery-aware GSPO to strengthen grounding and reduce hallucinations. Experiments demonstrate state-of-the-art detection accuracy and trustworthy, expert-aligned explanations.
format Preprint
id arxiv_https___arxiv_org_abs_2603_18577
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MedForge: Interpretable Medical Deepfake Detection via Forgery-aware Reasoning
Chen, Zhihui
He, Kai
Lei, Qingyuan
Pu, Bin
Zhang, Jian
Xu, Yuling
Feng, Mengling
Artificial Intelligence
Text-guided image editors can now manipulate authentic medical scans with high fidelity, enabling lesion implantation/removal that threatens clinical trust and safety. Existing defenses are inadequate for healthcare. Medical detectors are largely black-box, while MLLM-based explainers are typically post-hoc, lack medical expertise, and may hallucinate evidence on ambiguous cases. We present MedForge, a data-and-method solution for pre-hoc, evidence-grounded medical forgery detection. We introduce MedForge-90K, a large-scale benchmark of realistic lesion edits across 19 pathologies with expert-guided reasoning supervision via doctor inspection guidelines and gold edit locations. Building on it, MedForge-Reasoner performs localize-then-analyze reasoning, predicting suspicious regions before producing a verdict, and is further aligned with Forgery-aware GSPO to strengthen grounding and reduce hallucinations. Experiments demonstrate state-of-the-art detection accuracy and trustworthy, expert-aligned explanations.
title MedForge: Interpretable Medical Deepfake Detection via Forgery-aware Reasoning
topic Artificial Intelligence
url https://arxiv.org/abs/2603.18577