Research map / Label-efficient and robust learning / Semi-supervised learning
Beyond Binary Classification: A Semi-supervised Approach to Generalized AI-generated Image Detection
Hong-Hanh Nguyen-Le, Van-Tuan Tran, Thuc D. Nguyen, Nhien-An Le-Khac
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add assignment loss (Sinkhorn clustering) to binary loss (ablation) | TriDetect · Chameleon | AUC: 0.8033 -> 0.8560 | Improved |
| add consistency loss on top of assignment loss (ablation) | TriDetect · Chameleon | AUC: 0.8560 -> 0.8935 | Improved |
| add consistency loss on top of assignment loss (ablation) | TriDetect · Chameleon | ACC: 0.6335 -> 0.6258 vs binary only | Hurt |
| add both clustering losses (ablation) | TriDetect · Chameleon | EER: 0.2713 -> 0.1843 | Improved |
| set beta to 0.7 rather than 0.3 (ablation) | TriDetect · Chameleon | AUC: 0.8423 -> 0.8935 | Improved |
| set beta to 0.9 rather than 0.7 (ablation) | TriDetect · Chameleon | AUC: 0.8935 -> 0.8774 | Hurt |
| TriDetect vs Effort on WildFake | TriDetect · WildFake | avg ACC: 0.7522 -> 0.8254 | Improved |
| TriDetect vs Effort on AIGCDetectBenchmark | TriDetect · AIGCDetectBenchmark | avg AUC: 0.9783 -> 0.9869 | Improved |
Closest papers by meaning
- Community Forensics: Using Thousands of Generators to Train Fake Image Detectors (CVPR 2025 · similarity 0.89)
- A Bias-Free Training Paradigm for More General AI-generated Image Detection (CVPR 2025 · similarity 0.89)
- Breaking Latent Prior Bias in Detectors for Generalizable AIGC Image Detection (NeurIPS 2025 · similarity 0.88)
- Breaking Semantic Artifacts for Generalized AI-generated Image Detection (NeurIPS 2024 · similarity 0.87)
- Seeing What Matters: Generalizable AI-generated Video Detection with Forensic-Oriented Augmentation (NeurIPS 2025 · similarity 0.87)
- Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection (CVPR 2024 · similarity 0.87)
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.
