AI Paper Atlas

Research map / Video / Action recognition

GEM-TFL: Bridging Weak and Full Supervision for Forgery Localization through EM-Guided Decomposition and Temporal Refinement

Xiaodong Zhu, Yuanming Zheng, Suting Wang, Junqi Yang, Yuhong Yang, Weiping Tu, Zhongyuan Wang

CVPR 2026 · Video › Action recognition · approach: Weakly supervised temporal action localization · 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 Latent Attribute Decomposition (EM) to base MIL model (ablation)GEM-TFL base · AV-Deepfake1Maverage mAP: 5.12 -> 23.8Improved
add Temporal Consistency Refinement (ablation)GEM-TFL +LAD · AV-Deepfake1Maverage mAP: 23.8 -> 27.3Improved
add Graph-based Proposal Refinement (ablation)GEM-TFL +LAD+TCR · AV-Deepfake1Maverage mAP: 27.3 -> 31.9Improved
add Localization Phase regression (UMMAFormer) (ablation)GEM-TFL +LAD+TCR+GPR · AV-Deepfake1Maverage mAP: 31.9 -> 42.7Improved
use m=3 latent attributes instead of binary classification (ablation)LAD · AV-Deepfake1M classificationAUC: 98.70 -> 99.07Improved
use m=1 or 2 latent attributes (ablation)LAD · AV-Deepfake1M classificationAUC: 98.70 -> 98.58 (m=1), 98.48 (m=2)Hurt
use m>3 latent attributes (ablation)LAD · AV-Deepfake1M classificationAUC: 99.07 -> 98.90 (m=4)Hurt
GEM-TFL vs best weak baseline PseudoFormer on LAV-DFWS-TFL · LAV-DFaverage mAP: 64.7 -> 77.6Improved
GEM-TFL vs WMMT on AV-Deepfake1M (WMMT uses richer quadruple labels)WS-TFL · AV-Deepfake1Maverage mAP: 34.3 -> 42.7Improved
GEM-TFL vs fully supervised MFMSWS-TFL · AV-Deepfake1Maverage mAP: 42.7 vs 94.1Hurt

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