Research map / Image generation / Autoregressive and latent generation
SAIDO: Generalizable Detection of AI-Generated Images via Scene-Aware and Importance-Guided Dynamic Optimization in Continual Learning
Yongkang Hu, Yu Cheng, Yushuo Zhang, Yuan Xie, Zhaoxia Yin
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 IDOM on top of SAEM instead of no IDOM (ablation) | SAIDO · Protocol 1 | AA: 88.85 -> 95.61 | Improved |
| Add IDOM on top of SAEM instead of no IDOM (ablation) | SAIDO · Protocol 1 | AF: 11.41 -> 3.94 | Improved |
| Add SAEM scene experts on top of IDOM (ablation) | SAIDO · Protocol 1 | AA: 94.81 -> 95.61 | Improved |
| Replace RegO gradient projection (RAO) with IDOM (ablation) | SAIDO · Protocol 1 | AA: 92.89 -> 95.61 | Improved |
| Replace RegO gradient projection (RAO) with IDOM (ablation) | SAIDO · Protocol 1 | AF: 6.88 -> 3.94 | Improved |
| SAEM scene experts when scene distributions are concentrated (ablation) | IDOM-only model · Protocol 1 individual tasks | AA | Mixed |
| SAIDO vs best prior continual method without replay | Tang et al. / RegO · Protocol 1 | AF: 6.63 -> 3.94 | Improved |
| SAIDO vs second-best on open-world generators | Tang et al. · Open-world OOD | AA: 81.88 -> 91.35 | Improved |
| Gaussian noise degradation on SAIDO | SAIDO · Protocol 1 generators | AA1: 93.24 -> 79.61 | Hurt |
| JPEG compression degradation on SAIDO | SAIDO · Protocol 1 generators | AA1: 93.24 -> 92.27 | Hurt |
Closest papers by meaning
- Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection (ICML 2026 · similarity 0.88)
- Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes (CVPR 2026 · similarity 0.88)
- Breaking Semantic Artifacts for Generalized AI-generated Image Detection (NeurIPS 2024 · similarity 0.87)
- Adaptive Visual Scene Understanding: Incremental Scene Graph Generation (NeurIPS 2024 · similarity 0.87)
- CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AI (CVPR 2025 · similarity 0.85)
- A Bias-Free Training Paradigm for More General AI-generated Image Detection (CVPR 2025 · similarity 0.85)
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.
