Research map / Label-efficient and robust learning / Recognition and long-tail
When Sample Selection Bias Precipitates Model Collapse
Xinbao Qiao, Xianglong Du, Wei Liu, Jingqi Zhang, Peihua Mai, Meng Zhang, Yan Pang
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 |
|---|---|---|---|
| select synthetic data against a skewed local reference (Airplane class only) with CenterMatch | recursive DDPM training on CIFAR-10 · CIFAR-10, Airplane-only local reference | class diversity | Hurt |
| use local-reference selection baselines instead of random selection under non-IID reference | Accumulate-Subsample recursive DDPM · CIFAR-10, ExDir(1, 0.1) reference | FID: 106 -> 116 (CenterMatch), 106 -> 115 (CovMatch) | Hurt |
| use K-means selection instead of random | recursive DDPM · CIFAR-10, ExDir(1, 0.1) reference | FID: 106 -> 102 | No clear effect |
| use local-reference selection baselines in the IID partition | recursive DDPM · IID partition reference | FID | Improved |
| use collaborative geodesic interpolation proxy (Scheme I) instead of random selection | recursive DDPM · CIFAR-10, ExDir(1, 0.1) reference | FID: 106 -> 71 | Improved |
| use Wasserstein barycenter proxy (Scheme II) instead of random selection | recursive DDPM · STL-10, ExDir(1, 0.1) reference | FID: 95 -> 69 | Improved |
| use Scheme I instead of Scheme II | collaborative proxy selection · CIFAR-10, ExDir(1, 0.1) reference | FID: 85 -> 71 | Improved |
| use Scheme II (reusable barycenter) instead of Scheme I as number of clients grows | collaborative proxy selection · selection wall-clock time, K=100 | time (s): 49.2 -> 42.1 | Improved |
| add top-alpha selection toward a single local target to accumulate paradigm | Gaussian self-consuming loop · multivariate Gaussian simulation | Tr(Sigma_t)/Tr(Sigma_0) | Hurt |
| selection baselines on highly structured face data (CelebA) with a biased reference | CenterMatch/CovMatch/K-means · CelebA, ExDir(1, 0.1) reference | FID: 96 -> 87 (K-means, CenterMatch) | Improved |
Closest papers by meaning
- Collapse or Thrive: Perils and Promises of Synthetic Data in a Self-Generating World (ICML 2025 · similarity 0.85)
- Debiasing Synthetic Data Generated by Deep Generative Models (NeurIPS 2024 · similarity 0.84)
- Adaptive Data Optimization: Dynamic Sample Selection with Scaling Laws (ICLR 2025 · similarity 0.84)
- The Impact of Coreset Selection on Spurious Correlations and Group Robustness (NeurIPS 2025 · similarity 0.84)
- On the Burden of Achieving Fairness in Conformal Prediction (NeurIPS 2026 · similarity 0.84)
- Compute-Constrained Data Selection (ICLR 2025 · similarity 0.84)
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.
