AlignGemini: Generalizable AI-Generated Image Detection Through Task-Model Alignment
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908799509463040 |
|---|---|
| author | Chen, Ruoxin Gao, Jiahui Lin, Kaiqing Zhang, Keyue Zhao, Yandan Guan, Isabel Yao, Taiping Ding, Shouhong |
| author_facet | Chen, Ruoxin Gao, Jiahui Lin, Kaiqing Zhang, Keyue Zhao, Yandan Guan, Isabel Yao, Taiping Ding, Shouhong |
| contents | Vision Language Models (VLMs) are increasingly used for detecting AI-generated images (AIGI). However, converting VLMs into reliable detectors is resource-intensive, and the resulting models often suffer from hallucination and poor generalization. To investigate the root cause, we conduct an empirical analysis and identify two consistent behaviors. First, fine-tuning VLMs with semantic supervision improves semantic discrimination and generalizes well to unseen data. Second, fine-tuning VLMs with pixel-artifact supervision leads to weak generalization. These findings reveal a fundamental task-model misalignment. VLMs are optimized for high-level semantic reasoning and lack inductive bias toward low-level pixel artifacts. In contrast, conventional vision models effectively capture pixel-level artifacts but are less sensitive to semantic inconsistencies. This indicates that different models are naturally suited to different subtasks. Based on this insight, we formulate AIGI detection as two orthogonal subtasks: semantic consistency checking and pixel-artifact detection. Neglecting either subtask leads to systematic detection failures. We further propose the Task-Model Alignment principle and instantiate it in a two-branch detector, AlignGemini. The detector combines a VLM trained with pure semantic supervision and a vision model trained with pure pixel-artifact supervision. By enforcing clear specialization, each branch captures complementary cues. Experiments on in-the-wild benchmarks show that AlignGemini improves average accuracy by 9.5 percent using simplified training data. These results demonstrate that task-model alignment is an effective principle for generalizable AIGI detection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_06746 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AlignGemini: Generalizable AI-Generated Image Detection Through Task-Model Alignment Chen, Ruoxin Gao, Jiahui Lin, Kaiqing Zhang, Keyue Zhao, Yandan Guan, Isabel Yao, Taiping Ding, Shouhong Computer Vision and Pattern Recognition Artificial Intelligence Vision Language Models (VLMs) are increasingly used for detecting AI-generated images (AIGI). However, converting VLMs into reliable detectors is resource-intensive, and the resulting models often suffer from hallucination and poor generalization. To investigate the root cause, we conduct an empirical analysis and identify two consistent behaviors. First, fine-tuning VLMs with semantic supervision improves semantic discrimination and generalizes well to unseen data. Second, fine-tuning VLMs with pixel-artifact supervision leads to weak generalization. These findings reveal a fundamental task-model misalignment. VLMs are optimized for high-level semantic reasoning and lack inductive bias toward low-level pixel artifacts. In contrast, conventional vision models effectively capture pixel-level artifacts but are less sensitive to semantic inconsistencies. This indicates that different models are naturally suited to different subtasks. Based on this insight, we formulate AIGI detection as two orthogonal subtasks: semantic consistency checking and pixel-artifact detection. Neglecting either subtask leads to systematic detection failures. We further propose the Task-Model Alignment principle and instantiate it in a two-branch detector, AlignGemini. The detector combines a VLM trained with pure semantic supervision and a vision model trained with pure pixel-artifact supervision. By enforcing clear specialization, each branch captures complementary cues. Experiments on in-the-wild benchmarks show that AlignGemini improves average accuracy by 9.5 percent using simplified training data. These results demonstrate that task-model alignment is an effective principle for generalizable AIGI detection. |
| title | AlignGemini: Generalizable AI-Generated Image Detection Through Task-Model Alignment |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2512.06746 |