AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

MFVLR: Multi-Domain Fine-Grained Vision-Language Reconstruction for Generalizable Diffusion Face Forgery Detection and Localization

Yaning Zhang, Tianyi Wang, Zan Gao, Yibo Zhao, Chunjie Ma, Meng Wang

TPAMI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 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
use MFVLR instead of MFCLIPCLIP-based forgery detector · GenFace cross-forgery (EFS to AM)AUC: 78.76 -> 80.72Improved
use MFVLR instead of MFCLIPCLIP-based forgery detector · GenFace cross-forgery (AM to FS)AUC: 80.31 -> 83.40Improved
use MFVLR instead of MFCLIPCLIP-based forgery detector · GenFace cross-forgery (FS to AM)AUC: 80.38 -> 83.73Improved
use MFVLR instead of MSCCNetlocalization detector · GenFace cross-forgery (EFS to FS)AUC: 81.73 -> 84.03Improved
use MFVLR instead of MFCLIPCLIP-based forgery detector · GenFace cross-forgery (AM to EFS)ACC: 53.29 -> 56.73Improved
use MFVLR instead of SIDA for localizationdiffusion forgery localizer · GenFace cross-diffusion (DDPM to LatDiff)mIoU: 49.26 -> 59.34Improved
use MFVLR instead of SIDA for localizationdiffusion forgery localizer · GenFace cross-diffusion (LatDiff to DDPM)mIoU: 69.66 -> 74.89Improved
use MFVLR instead of MSCCNet for localizationdiffusion forgery localizer · GenFace cross-diffusion (CollDiff to LatDiff)mIoU: 89.57 -> 91.37Improved
use MFVLR instead of MFCLIP when trained on DDPMCLIP-based forgery detector · GenFace cross-diffusion (DDPM to DiffFace)ACC: 88.89 -> 82.65Hurt
use MFVLR instead of MSCCNetlocalization detector · GenFace cross-forgery (FS to AM)AUC: 80.56 -> 83.73Improved

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