AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Towards Generalizable AI-Generated Image Detection via Image-Adaptive Prompt Learning

Yiheng Li, Zichang Tan, Guoqing Xu, Zhen Lei, Xu Zhou, Yang Yang

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 5 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 MLP-based adapters (ablation)CLIP ViT-L/14 baseline · UniversalFakeDetectmAcc: 84.19 -> 89.02Improved
add learnable tokens (ablation)CLIP with adapters · UniversalFakeDetectmAcc: 89.02 -> 91.83Improved
add image-adaptive prompts (ablation)CLIP with adapters and tokens · UniversalFakeDetectmAcc: 91.83 -> 94.88Improved
add optimal input selection (ablation)IAPL · UniversalFakeDetectmAcc: 94.88 -> 95.61Improved
add general condition to forgery-specific condition (ablation)IAPL · UniversalFakeDetectmAcc: 92.38 -> 93.32Improved
add learnable scaling factor (ablation)IAPL · UniversalFakeDetectmAcc: 93.32 -> 93.51Improved
add test-time token tuning (ablation)IAPL · UniversalFakeDetectmAcc: 93.51 -> 94.58Improved
add confidence selection (ablation)IAPL · UniversalFakeDetectmAcc: 94.58 -> 94.88Improved
use averaged entropy instead of pointwise entropy loss (ablation)IAPL test-time tuning · UniversalFakeDetectmAcc: 95.35 -> 95.61Improved
increase test-time tuning steps T from 1 to 3 (ablation)IAPL · UniversalFakeDetectmAcc: 94.47 -> 94.94Improved
insert adapters evenly at [4,24,4] instead of first six layers (ablation)IAPL adapters · UniversalFakeDetectmAcc: 78.27 -> 89.02Improved
insert adapters into more layers ([18,24,1]) instead of six evenly (ablation)IAPL adapters · UniversalFakeDetectmAcc: 89.02 -> 87.74Hurt

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