AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

ReAlign: Generalizable Image Forgery Detection via Reasoning-Aligned Representation

Qing Huang, Zhipei Xu, Xuanyu Zhang, Xiangyu Yu, Jian Zhang

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 2 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
use reasoning text instead of image captions as alignment text (ablation)ReAlign · UltraSynth-10kmean accuracy: 97.09 -> 91.63 (-5.46%) when replaced by captionsImproved
add class-label prefix to reasoning text (ablation)ReAlign · UltraSynth-10kmean accuracy: 96.87 -> 97.09 (0.22% decrease without prefix)Improved
reasoning text alone versus caption alone (ablation)ReAlign · UltraSynth-10kmean accuracy: 88.32 -> 96.87Improved
reasoning text alone versus class label alone (ablation)ReAlign · UltraSynth-10kmean accuracy: 91.33 -> 96.87Improved
use joint optimization instead of sequential (LoRA) (ablation)ReAlign · UltraSynth-10kmean accuracy: 84.08 -> 97.09 (-13.01%)Improved
use joint optimization instead of sequential (full fine-tuning) (ablation)ReAlign · UltraSynth-10kmean accuracy: 79.07 -> 94.69Improved
use LoRA instead of full fine-tuning of image encoder (ablation)ReAlign · UltraSynth-10kmean accuracy: 94.69 -> 97.09 (-2.40%)Improved
add reasoning-alignment contrastive loss to frozen encoder (ablation)ReAlign · UltraSynth-10kmean accuracy: 89.15 -> 97.09 (-7.94%)Improved
add reasoning-alignment contrastive loss to LoRA encoder (ablation)ReAlign · UltraSynth-10kmean accuracy: 93.68 -> 97.09 (-3.41%)Improved
distill AIGI-R1 reasoning into CLIP (ReAlign vs AIGI-R1)ReAlign · AIGCDetectBenchmarkmean accuracy: 91.77 -> 96.14Improved
ReAlign versus AIDEReAlign · AIGI-Holmesmean accuracy: 97.00 -> 99.44Improved
ReAlign versus AIGI-R1 on unseen proprietary generatorsReAlign · UltraSynth-10kmean accuracy: 96.42 -> 97.09Mixed

Later papers that compared with this method

2 later accepted papers report ReAlign as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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