- Rating
- 3(+1)
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes SPARE, a CLIP based AI-generated image detector that exploits representation changes induced by patch shuffling. The authors argue that real images rely more heavily on global spatial coherence than generated images, resulting in stronger feature responses when spatial structure is disrupted. To exploit this observation, SPARE computes layer-wise feature differences between original and path shuffled image, applies adaptive gating, and combines these responses with original representations for classification.
Critical strengths
- The method is build upon a frozen CLIP backbone, avoiding expensive backbone fine-tuning.
- The paper evaluates generalization across multiple GAN and diffusion based generators.
- The authors include component ablations, perturbation comparisons, and robustness experiments.
Critical weaknesses
- SFLD (gye et al., 205) already introduced patch shuffling with frozen CLIP representations for AI-generated image detection, explicitly aiming to reduce semantic bias while preserving local texture cues. SPARE builds upon highly similar underlying motivation and methodological foundation, yet does not adequately establish its novelty relative to this work.
- The authors attribute stronger structural responses in real images to their greater dependence on physical and spatial coherence. However, the presented evidence does not establish this explanation or rule out alternative causes.
- Table 5, gaussian noise achieves an average accuracy of 92.82 compared with 93.81 for patch shuffle. This relatively small difference raises questions about whether the reported improvements specifically arise from disrupting global structure.
- The paper does not convincingly establish that this combination constitutes a sufficiently substantial methodological contribution.
Questions for authors
See weaknesses
7 comments
Response to Reviewer DcnY (Part 2/4)
W2. Evidence for the physical-coherence explanation.
About physical and spatial coherence. We clarify that physical constraints provide one possible explanation for the different structural responses observed in Fig. 1. As discussed by Garces et al. (2022), image appearance is jointly influenced by illumination, material properties, and geometry. We hypothesize that these factors induce spatial dependencies among different regions in real images, making their global representations more sensitive to structural perturbations. Nevertheless, SPARE focuses on capturing and exploiting these representation differences rather than identifying their underlying causes.
About empirical evidence. To investigate potential alternative explanations, we conduct additional experiments to examine the effects of image complexity, texture, and compression history on structural responses. Specifically, we evaluate BPP and Sobel gradient magnitude as proxies for image complexity and texture, respectively.
- Token-aligned shuffling. Because the patches coincide with ViT tokens, every token keeps exactly the same pixels and only its position changes. The response therefore contains no seam artifacts and no change in local texture within any token.
- Complexity and texture point the other way. For each image we compute JPEG compressibility (bits per pixel after re-encoding at Q = 75) and texture strength (mean grayscale Sobel gradient magnitude). Generated images are slightly more complex and textured than real ones (ProGAN: bpp 1.691 vs. 1.663, Sobel 84.18 vs. 83.14; SD v1.4: bpp 1.808 vs. 1.722, Sobel 85.54 vs. 82.93), yet real images respond more strongly. If merely tracked complexity, the opposite would hold.
- Controls do not reduce the separation. We (i) stratify images into five bins by bpp or by Sobel and compute the sample-weighted within-bin AUROC, (ii) regress on complexity (quadratic in log-bpp and Sobel with interaction, pooled over both classes) and use the residual as the score, and (iii) re-encode all images with JPEG Q = 95 to align compression history. AUROC of , with real images as the positive class:
| Setting | Raw (95% CI) | Stratified by bpp | Stratified by Sobel | Residualized |
|---|---|---|---|---|
| ProGAN | 0.7975 (0.7802–0.8198) | 0.8187 | 0.8243 | 0.8243 |
| ProGAN, JPEG Q95 | 0.7902 (0.7676–0.8125) | 0.8109 | 0.8166 | 0.8181 |
| SD v1.4 | 0.6150 (0.5876–0.6373) | 0.6043 | 0.6206 | 0.6557 |
| SD v1.4, JPEG Q95 | 0.6015 (0.5742–0.6237) | 0.6104 | 0.6378 | 0.6208 |
Overall, these experiments suggest that the observed structural response differences cannot be simply explained by image complexity, texture intensity, or compression history. Although they do not establish physical and spatial coherence as the sole underlying cause, they provide further empirical support for the discriminative value of structural responses. We have included these additional experiments in the revised manuscript.
Reference:
Garces, E., Rodriguez-Pardo, C.,·Casas, D.,·Lopez-Moreno, J. A Survey on Intrinsic ImagesDelving: Deep Into Lambert and Beyond. International Journal of Computer Vision, 130(3):836–868, 2022.
Response to Reviewer DcnY (Part 3/4)
W3. Gaussian noise vs. Patch Shuffle (Table 5).
To further clarify the effectiveness of Patch Shuffle, we conduct additional ablation experiments to distinguish the benefits of structural response modeling from those of simply introducing perturbed images:
| Perturbation | ForenSynths | GenImage | Average |
|---|---|---|---|
| Gaussian Noise (Static shuffled feature ) | 89.13 | 81.63 | 85.38 |
| Gaussian Noise (Structural Response ) | 95.56 | 90.08 | 92.82 |
| Patch Mask (Static shuffled feature ) | 64.40 | 55.05 | 59.73 |
| Patch Mask (Structural Response ) | 88.25 | 86.77 | 87.51 |
| Patch Shuffle (Static shuffled feature ) | 94.74 | 87.62 | 91.18 |
| Patch Shuffle (Structural Response ) | 95.66 | 91.96 | 93.81 |
(a) Does the response formulation matter?
Yes. For every perturbation, classifying the response outperforms classifying the perturbed view, with the original branch and all other settings unchanged.
(b) Does the type of perturbation matter?
Yes. When the perturbed view itself must carry the signal, Patch Shuffle outperforms Gaussian noise by 5.80 pp and Patch Mask by 31.45 pp, so the perturbation type strongly affects how much discriminative information the perturbed view holds. Under the response formulation the gap narrows to 0.99 pp.
(c) Positioning.
We do not claim that only structural perturbations carry information. RIGID (He et al., 2024) showed that sensitivity to small noise in foundation-model features is informative, and our Gaussian-noise results are consistent with this. Our claim is that the representation response is an effective cue, and that Patch Shuffle is its most effective instance among the perturbations we tested. It is also the perturbation that best isolates dependence on spatial arrangement: with token-aligned patches the content of every token is unchanged and only positions move, whereas noise and masking alter the content itself.
Reference:
Z. He, P.-Y. Chen, T.-Y. Ho. RIGID: A Training-Free and Model-Agnostic Framework for Robust AI-Generated Image Detection. arXiv:2405.20112, 2024.
Response to Reviewer DcnY (Part 1/4)
We thank the reviewer for the careful review, and in particular for pointing us to SFLD, which we should have discussed. We also appreciate the recognition of SPARE's efficiency with a frozen backbone, its cross-generator evaluation, and its ablation and robustness experiments. We address each point below.
W1. Relation to SFLD.
We agree that SFLD (Gye et al., WACV 2025) is closely related and should have been discussed. Both methods apply patch shuffling with a frozen CLIP ViT-L/14, and both aim to reduce content bias. We also correct an inaccurate sentence in our introduction stating that patch shuffling had previously been used only as a training-time augmentation; SFLD uses shuffled views at inference as well. The revision discusses SFLD in the Introduction, Related Work and Experiments.
The two methods differ in which signal is classified:
SFLD trains an independent linear classifier on the final-layer CLIP feature of a single view: the original image, or a shuffled image at patch size 56 or 28. At test time it averages 10 shuffled views per patch size and averages the logits of the three classifiers. Each classifier separates real from fake from one view alone; views are combined only by averaging scores.
SPARE classifies the change between the two views, δℓ=Fclsℓ−F~clsℓ, over a window of intermediate layers, jointly with the original representation. Its input is how the representation reacts to the shuffle, not what either view looks like.
This difference is testable. Keeping the original branch, architecture, layers and training data unchanged, we replaced the response δ with the static feature of the shuffled image F~, so both variants receive exactly the same two views:
Given the same two views, modeling their difference explicitly improves mean ACC by 2.63 pp, and by 4.34 pp on the cross-dataset GenImage benchmark.
SPARE is also cheaper at inference: 2 CLIP forward passes per image, compared with 21 for SFLD (1 original view + 2 patch sizes × 10 shuffled views).
Reference:
Seoyeon Gye, Junwon Ko, Hyounguk Shon, Minchan Kwon, and Junmo Kim . Sfld: Reducing thecontent bias for ai-generated image detection. In WACV, 2025 .