AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Uncovering and Mitigating Destructive Multi-Embedding Attacks in Deepfake Proactive Forensics

Lixin Jia, Haiyang Sun, Zhiqing Guo, Yunfeng Diao, Dan Ma, Gaobo Yang

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Neural network watermarking · 4 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
Apply a second watermark embedding (MEA)existing proactive forensics methods · CelebA-HQBER (%): 0.50 -> 38.17Hurt
Add AIS fine-tuning (simulate MEA with resilience loss) (ablation)SepMark · CelebA-HQ, 2 embeddingsBER (%): 48.6223 -> 0.0465Improved
Add AIS fine-tuning (ablation)HiDDeN · CelebA-HQ, 2 embeddingsBER (%): 49.5370 -> 2.0679Improved
Add AIS fine-tuning (ablation)EditGuard · CelebA-HQ, 5 embeddingsBER (%): 34.7820 -> 2.2408Improved
Raise AIS loss weight lambda_A from 0 to 10 (ablation)LampMark · CelebA-HQ, 5 embeddingsBER (%): 18.60 -> 0.22Improved
Raise AIS loss weight lambda_A from 0 to 10 (ablation)LampMark · CelebA-HQ, 5 embeddingsPSNR (dB): 17.93 -> 20.32Hurt
Add AIS fine-tuning (ablation)HiDDeN · CelebA-HQ, 1 embeddingPSNR (dB): 25.2963 -> 22.0898Hurt
Add AIS fine-tuning (ablation)LampMark · CelebA-HQ, 1 embeddingPSNR (dB): 27.0208 -> 28.6729Improved
Add AIS fine-tuning, evaluated cross-dataset (ablation)SepMark · LFW, 2 embeddingsBER (%): 49.27 -> 20.16Improved
Add AIS fine-tuning against cross-model MEA (ablation)EditGuard · Cross-model MEA, SepMark attacker n=1BER (%): 50.51% -> 23.17%Improved
Add AIS fine-tuning against cross-model MEA (ablation)LampMark defending vs SepMark attacker · Cross-model MEABER (%): 11.05% -> 10.04%Mixed

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