Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.19430 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910002073042944 |
|---|---|
| author | Xiao, Yao Chen, Weiyan Chen, Jiahao Cao, Zijie Deng, Weijian Yang, Binbin Dong, Ziyi Ji, Xiangyang Ke, Wei Wei, Pengxu Lin, Liang |
| author_facet | Xiao, Yao Chen, Weiyan Chen, Jiahao Cao, Zijie Deng, Weijian Yang, Binbin Dong, Ziyi Ji, Xiangyang Ke, Wei Wei, Pengxu Lin, Liang |
| contents | Current AI-Generated Image (AIGI) detection approaches predominantly rely on binary classification to distinguish real from synthetic images, often lacking interpretable or convincing evidence to substantiate their decisions. This limitation stems from existing AIGI detection benchmarks, which, despite featuring a broad collection of synthetic images, remain restricted in their coverage of artifact diversity and lack detailed, localized annotations. To bridge this gap, we introduce a fine-grained benchmark towards eXplainable AI-Generated image Detection, named X-AIGD, which provides pixel-level, categorized annotations of perceptual artifacts, spanning low-level distortions, high-level semantics, and cognitive-level counterfactuals. These comprehensive annotations facilitate fine-grained interpretability evaluation and deeper insight into model decision-making processes. Our extensive investigation using X-AIGD provides several key insights: (1) Existing AIGI detectors demonstrate negligible reliance on perceptual artifacts, even at the most basic distortion level. (2) While AIGI detectors can be trained to identify specific artifacts, they still substantially base their judgment on uninterpretable features. (3) Explicitly aligning model attention with artifact regions can increase the interpretability and generalization of detectors. The data and code are available at: https://github.com/Coxy7/X-AIGD. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_19430 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Unveiling Perceptual Artifacts: A Fine-Grained Benchmark for Interpretable AI-Generated Image Detection Xiao, Yao Chen, Weiyan Chen, Jiahao Cao, Zijie Deng, Weijian Yang, Binbin Dong, Ziyi Ji, Xiangyang Ke, Wei Wei, Pengxu Lin, Liang Computer Vision and Pattern Recognition Current AI-Generated Image (AIGI) detection approaches predominantly rely on binary classification to distinguish real from synthetic images, often lacking interpretable or convincing evidence to substantiate their decisions. This limitation stems from existing AIGI detection benchmarks, which, despite featuring a broad collection of synthetic images, remain restricted in their coverage of artifact diversity and lack detailed, localized annotations. To bridge this gap, we introduce a fine-grained benchmark towards eXplainable AI-Generated image Detection, named X-AIGD, which provides pixel-level, categorized annotations of perceptual artifacts, spanning low-level distortions, high-level semantics, and cognitive-level counterfactuals. These comprehensive annotations facilitate fine-grained interpretability evaluation and deeper insight into model decision-making processes. Our extensive investigation using X-AIGD provides several key insights: (1) Existing AIGI detectors demonstrate negligible reliance on perceptual artifacts, even at the most basic distortion level. (2) While AIGI detectors can be trained to identify specific artifacts, they still substantially base their judgment on uninterpretable features. (3) Explicitly aligning model attention with artifact regions can increase the interpretability and generalization of detectors. The data and code are available at: https://github.com/Coxy7/X-AIGD. |
| title | Unveiling Perceptual Artifacts: A Fine-Grained Benchmark for Interpretable AI-Generated Image Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.19430 |