AI Paper Atlas

Research map / Video / Video-language understanding

AvatarShield: Visual Reinforcement Learning for Human-Centric Synthetic Video Detection

IJCV 2026 · Video › Video-language understanding · approach: Event and anomaly reasoning in video

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 residual extractor (ablation)AvatarShield · FakeHumanVid mean (in-domain)AUC: 0.7989 -> 0.9381Improved
use residuals instead of raw VAE reconstructions (ablation)AvatarShield · FakeHumanVid mean (in-domain)AUC: 0.8609 -> 0.9381Improved
add temporal compensation reward (ablation)AvatarShield · FakeHumanVid mean (in-domain)AUC: 0.8451 -> 0.9381Improved
train with GRPO instead of SFT (ablation)AvatarShield · FakeHumanVid mean (in-domain)AUC: 0.8301 -> 0.9381Improved
add temporal compensation reward (ablation)AvatarShield · FakeHumanVid StableAnimator (in-domain)AUC: 0.7522 -> 0.9573Improved
add temporal compensation reward (ablation)AvatarShield · FakeHumanVid CogVideo (in-domain)AUC: 0.8210 -> 0.9571Improved
add residual extractor (ablation)AvatarShield · FakeHumanVid StableAnimator (in-domain)AUC: 0.7176 -> 0.9573Improved
AvatarShield instead of MM-DetAvatarShield · FakeHumanVid mean (in-domain)AUC: 0.8979 -> 0.9381Improved
AvatarShield instead of MM-Det, trained on pose dataAvatarShield · FakeHumanVid Hallo3 (cross-domain, trained on pose)AUC: 0.7505 -> 0.8592Improved
AvatarShield instead of MM-Det, trained on text dataAvatarShield · FakeHumanVid Wanx (cross-domain, trained on text)AUC: 0.7999 -> 0.9134Improved
AvatarShield instead of MM-Det, trained on mixed dataAvatarShield · FakeHumanVid cross-domain mean (trained on mixed)AUC: 0.8114 -> 0.8783Improved
AvatarShield instead of RECCEAvatarShield · FakeHumanVid audio-driven mean (Hallo3, HelloMeme)AUC: 0.8601 -> 0.9353Improved

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