- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes CRAS, a black-box anti-forensic framework for AI-generated image detectors. The method combines cross-modal reconstruction to manipulate semantic cues with artifact suppression to reduce synthesis-specific traces. Experiments across four datasets and ten detectors demonstrate strong attack effectiveness and high visual fidelity, supported by comparisons with existing methods and ablation studies.
Critical strengths
-
Meaningful research problem. The paper investigates the robustness of AI-generated image detectors by examining their reliance on semantic and synthesis-specific artifact cues. This offers a useful perspective on understanding detector vulnerabilities.
-
Comprehensive experimental evaluation. The method is evaluated across four datasets and ten detectors, covering both CLIP-based and conventional architectures. The results demonstrate strong attack effectiveness while maintaining good image quality, supporting the practical value of the proposed framework.
Critical weaknesses
-
Insufficient validation of the underlying mechanism. The paper claims that CRM and ASM independently manipulate semantic and artifact cues, respectively. However, diffusion-based reconstruction may also alter low-level artifacts, while artifact suppression may affect semantic representations. The current ablation studies demonstrate the effectiveness of each module but do not sufficiently establish the claimed separation of these two mechanisms.
-
Limited methodological novelty. Both diffusion-based latent optimization with CLIP guidance and encoder-decoder-based artifact suppression build on established techniques. The main contribution appears to be their combination, while the claimed complementary mechanisms are not rigorously validated. This weakens the novelty and contribution to understanding detector vulnerabilities.
Questions for authors
-
Mechanism validation (Clarification). How can the authors establish that CRM primarily manipulates semantic cues while ASM specifically suppresses synthesis artifacts? Could the authors provide quantitative evidence showing how each module affects both types of features?
-
Contribution of semantic optimization (Clarification). How much of CRM's attack effectiveness comes from the semantic alignment objective rather than the diffusion reconstruction process itself? A comparison with standard DDIM reconstruction without semantic optimization would help clarify the contribution of the proposed mechanism.
2 comments
Response to Q2
We separated CRM into two components, reconstruction and CLIP-guided semantic alignment, and extended the original ablation study with additional experiments.
Table 3: Ablation study on the GenImage dataset. Results are reported as detection accuracy. ✓ and × indicate that a component is enabled or disabled, respectively. When all three components are disabled, evaluation is performed on the original fake images. The CRM module is enabled when both Reconstruction and Semantic Alignment are marked ✓.
| Reconstruction Only | Semantic Alignment | ASM | UniFD | DGS-Net | Effort | CLIP+Linear | CLIP+LoRA | CNNSpot | GramNet | ResNet50 | SAFE | NPR | Avg. |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ✘ | ✘ | ✘ | 0.70 | 0.61 | 0.87 | 0.64 | 0.69 | 0.43 | 0.54 | 0.33 | 0.70 | 0.69 | 0.62 |
| ✔ | ✘ | ✘ | 0.89 | 0.74 | 0.92 | 0.78 | 0.94 | 0.61 | 0.85 | 0.63 | 0.94 | 0.99 | 0.82 |
| ✔ | ✔ | ✘ | 0.41 | 0.34 | 0.54 | 0.20 | 0.12 | 0.33 | 0.32 | 0.00 | 0.29 | 0.38 | 0.33 |
| ✔ | ✘ | ✔ | 0.75 | 0.98 | 0.87 | 0.38 | 0.64 | 0.21 | 0.44 | 0.44 | 0.92 | 0.92 | 0.65 |
| ✘ | ✘ | ✔ | 0.64 | 0.58 | 0.77 | 0.40 | 0.54 | 0.16 | 0.38 | 0.31 | 0.60 | 0.67 | 0.50 |
| ✔ | ✔ | ✔ | 0.33 | 0.33 | 0.22 | 0.05 | 0.03 | 0.08 | 0.22 | 0.00 | 0.20 | 0.37 | 0.18 |
Analysis: Reconstruction without semantic optimization makes fake images easier to detect, while adding semantic alignment substantially improves attack effectiveness.
We thank the reviewer for the feedback and have conducted additional experiments to address your questions.
Response to Weakness 1 & Q1
To quantify the effects of each module on semantic and artifact cues, we use the following methods:
Table 1: Effect of Each Module on Semantic Cues.
Analysis: The greater distance between the features of CRM-processed images and those of the original images indicates that CRM primarily modifies semantic cues.
Table 2: Effect of Each Module on Artifact Cues.
Analysis: The mid- and high-frequency power spectra of ASM-processed images are closer to those of the original images, indicating that ASM suppresses more artifacts.
Response to Weakness 2
We explain how CRM and ASM each differ from existing methods as follows: