Research map / Image generation / Autoregressive and latent generation
Rethinking Bias in Generative Data Augmentation for Medical AI: A Frequency Recalibration Method
Chi Liu, Jincheng Liu, Congcong Zhu, Minghao Wang, Sheng Shen, Jia Gu, Tianqing Zhu, Wanlei Zhou
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 generative data augmentation (GDA) to raw training | ResNet50 · Cardiomegaly | accuracy: 0.792 -> 0.782 | Hurt |
| add GDA to raw training | DenseNet · Brain tumor | accuracy: 0.742 -> 0.732 | Hurt |
| add GDA to raw training | ResNet50 · Brain tumor | accuracy: 0.783 -> 0.723 | Hurt |
| add FreRec calibration to GDA | ResNet50 · Brain tumor | accuracy: 0.723 -> 0.843 | Improved |
| add FreRec calibration to GDA | ResNet50 · Cardiomegaly | accuracy: 0.782 -> 0.832 | Improved |
| add FreRec calibration to GDA | DenseNet · DR | accuracy: 0.813 -> 0.834 | Improved |
| replace FreRec with DoGE domain adaptation | ViT-B-16 · Cardiomegaly | accuracy: 0.838 vs 0.818 | No clear effect |
| replace FreRec with DoGE domain adaptation | DenseNet · DR | accuracy: 0.834 vs 0.841 | Mixed |
| add RHM after SHR (SHR only vs full FreRec) (ablation) | FreRec · Cardiomegaly | accuracy: 0.79 -> 0.84 | Improved |
| add RHM after SHR (SHR only vs full FreRec) (ablation) | FreRec · Brain tumor | accuracy: 0.73 -> 0.81 | Improved |
| add SHR before RHM (ablation) | FreRec · Cardiomegaly | accuracy: 0.82 -> 0.84 | Improved |
| add FESA to RHM (ablation) | FreRec · Brain tumor | accuracy: 0.75 -> 0.80 | Improved |
Closest papers by meaning
- Fourier-basis Functions to Bridge Augmentation Gap: Rethinking Frequency Augmentation in Image Classification (CVPR 2024 · similarity 0.82)
- Dual Data Alignment Makes AI-Generated Image Detector Easier Generalizable (NeurIPS 2025 · similarity 0.82)
- MMed-RAG: Versatile Multimodal RAG System for Medical Vision Language Models (ICLR 2025 · similarity 0.81)
- AIF-SFDA: Autonomous Information Filter Driven Source-Free Domain Adaptation for Medical Image Segmentation (AAAI 2025 · similarity 0.81)
- Instruct Where the Model Fails: Generative Data Augmentation via Guided Self-contrastive Fine-tuning (AAAI 2025 · similarity 0.81)
- Large-scale and Fine-grained Vision-language Pre-training for Enhanced CT Image Understanding (ICLR 2025 · similarity 0.81)
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.
