Research map / Probabilistic and scientific modeling / Variational and generative sampling
Assessing Generative Models via Precision and Recall
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 target-distribution classes to the generated samples | Class-based sample distributions evaluated with PRD · CIFAR-10 | recall | Helped |
| Add classes absent from the target distribution to generated samples | Class-based sample distributions evaluated with PRD · CIFAR-10 | precision | Hurt |
| Drop target-distribution classes from generated samples | Class-based sample distributions evaluated with PRD · CIFAR-10 | recall | Hurt |
| Use GANs rather than VAEs | Generative models evaluated on Fashion-MNIST · Fashion-MNIST | precision | Mixed |
| Use VAEs rather than GANs | Generative models evaluated on Fashion-MNIST · Fashion-MNIST | recall | Helped |
| Evaluate with a two-dimensional precision-recall description instead of a single symmetric score | Generative-model comparisons · MNIST and CelebA | failure-case discrimination: 32/29 for MNIST and 65/62 for CelebA | Helped |
Closest papers by meaning
- Precision-Recall Divergence Optimization for Generative Modeling with GANs and Normalizing Flows (NeurIPS 2023 · similarity 0.92)
- Emergent Asymmetry of Precision and Recall for Measuring Fidelity and Diversity of Generative Models in High Dimensions (ICML 2023 · similarity 0.91)
- A Unifying Information-theoretic Perspective on Evaluating Generative Models (AAAI 2025 · similarity 0.90)
- MIND: Monge Inception Distance for Generative Models Evaluation (NeurIPS 2026 · similarity 0.89)
- Statistically Optimal Generative Modeling with Maximum Deviation from the Empirical Distribution (ICML 2024 · similarity 0.88)
- Efficient Precision and Recall Metrics for Assessing Generative Models using Hubness-aware Sampling (ICML 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.