AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Attributing Image Generative Models using Latent Fingerprints

Guangyu Nie, Changhoon Kim, Yezhou Yang, Yi Ren

ICML 2023 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 3 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
Constrain alpha estimate (alpha-regularization) during attribution (ablation)latent fingerprint attribution, StyleGAN2 FFHQ · FFHQ SG2, no postprocessattribution accuracy: 0.877 -> 0.983Improved
Constrain alpha estimate (alpha-regularization) during attribution (ablation)latent fingerprint attribution, LDM FFHQ · FFHQ LDM, no postprocessattribution accuracy: 0.364 -> 0.996Improved
Add Latin hypercube parallel search with multiple initial guesses for alpha (ablation)latent fingerprint attribution · FFHQ SG2 and LDM, no postprocessattribution accuracy: SG2 0.711 -> 0.983; LDM 0.872 -> 0.996Improved
Move fingerprint dimensions from major to minor principal components of Sigma_w (LDM, sigma=1) (ablation)latent fingerprints, LDM · FFHQ LDM, sigma=1.0attribution accuracy: 0.99 -> 0.00 (PC[0:64] to PC[4000:4064])Hurt
Move fingerprint dimensions from major to minor principal components of Sigma_w (LDM, sigma=1) (ablation)latent fingerprints, LDM · FFHQ LDM, sigma=1.0FID: 33.62 -> 12.77 (lower is better)Improved
Increase fingerprint strength sigma from 1 to 6 on major PCs (SG2) (ablation)latent fingerprints, StyleGAN2 PC[0:64] · FFHQ SG2FID: 129.0 at sigma 0.6 vs 101.3 at sigma 6.0 (mixed direction); accuracy limited benefit beyond small sigmaHurt
Increase key length d_phi from 32 to 128 bits (ablation)latent fingerprints, StyleGAN2 · FFHQ SG2attribution accuracy: 0.982 -> 0.973 (32 to 128 bits)Hurt
Increase key length d_phi from 32 to 128 bits (ablation)latent fingerprints, StyleGAN2 · FFHQ SG2FID: 8.49 -> 9.61 (32 to 128 bits)Hurt
Use latent fingerprints (PC[32:40], 8-bit) instead of shallow fingerprints under Combo postprocessStyleGAN2 FFHQ · Combo postprocess, unknown attackattribution accuracy: 0.65 (ours, UK) vs 0.83 (BL, UK); 0.89 vs 0.88 with KNMixed
Use latent fingerprints (PC[32:40]) instead of shallow fingerprintsStyleGAN2 FFHQ · Combo postprocessFID: 12.35 vs 100.15 (lower is better)Improved
Use latent fingerprints (PC[32:40]) instead of shallow fingerprintsStyleGAN2 FFHQ · Combo postprocessSSIM (secrecy): 0.73 vs 0.66Improved
Postprocess-specific modified LPIPS for attribution (known-attack setting) (ablation)latent fingerprint attribution · Combo postprocess, KNattribution accuracy: PC[32:40] UK 0.65 -> KN 0.89Improved

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