AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

PP-Mark: Provable and Publicly Verifiable Watermarking for Generative AI

HANKYEONG KO, Taehoon Kim

NeurIPS 2026 · Adversarial robustness and security › Black-box and transfer attacks · approach: Data-free substitute training · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes PP-Mark, a provenance framework that ties a public, lightweight statistical watermark detector to a zero-knowledge proof of the embedding process so that generated content can be publicly verified without enabling forgery or adversarial removal.

Problem

Publicly verifiable provenance for generative AI images is needed, but existing watermarks become vulnerable to forgeries and adversarial optimization attacks once their detectors are made public.

Approach

Anchors lightweight statistical watermark detection to a zero-knowledge proof of the embedding process, so the public can verify that content was watermarked correctly without the detector being exploitable for forgery or removal.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ZKP receipt verification (accept mode) on top of score (ablation)PP-Mark · White-box PGD forgeryFAR@1%FPR: 0.76 -> 0.00Improved
add ZKP receipt verification on top of signature+score (ablation)Sig+Score · Forged binding attackFAR: 1.00 -> 0.00Improved
bind watermark to secret key via context-bound latent embedding (score only) (ablation)PP-Mark (score) · Forged binding attackFAR: 1.00 -> <=0.01Improved
add score check to signature-only verification (ablation)Sig-only · Compliance bypass attackFAR: 1.00 -> 0.00Improved
add score check to signature-only verification (ablation)Sig-only · Forged binding attackFAR: 1.00 -> 1.00No clear effect
use PP-Mark score detector instead of other detectorsPP-Mark (score) · Imprint forgery transfer, Tree-Ring sourceFAR@1%FPR: 0.0400 -> 0.0033Improved
use PP-Mark score detector instead of WIND under imprint forgery from GSPP-Mark (score) · Imprint forgery transfer, GS sourceFAR@1%FPR: 0.0100 -> 0.0067No clear effect
increase watermark strength alpha from 4.0 to 4.5 (ablation)PP-Mark · SD2.1 image qualityBRISQUE p95: 34.9 -> 40.5Hurt
increase number of sampled positions N from 200 to 1000 (ablation)PP-Mark · Score variancescore std across images: 0.1409 -> 0.1301Improved
increase N from 1000 to 1500 (ablation)PP-Mark · Score variancescore std across images: 3.0% additional reductionNo clear effect
increase N from 200 to 1000 (SP1 v5) (ablation)PP-Mark proof · SP1 proving costprove time (s): 32.309 -> 48.572Hurt
switch SP1 v5 to v6.3.1 circuit (ablation)PP-Mark proof · N=1000 provingprove time (s): 48.572 -> 17.0Improved

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