Research map / Probabilistic and scientific modeling / Variational and generative sampling
Flow-based density estimation
Applies flows to density-ratio estimation and likelihood analysis with autoregressive structure.
29 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in Variational and generative sampling, keyword cluster “flows · normalizing · invertible”
State History Tried Checklist Preprints Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
9 papers in 2023–24 and 5 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 2.2% to 1.0%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2017 · Masked Autoregressive Flow for Density Estimation (NeurIPS 2017 · 201 citations)
- 2018 · Deep Density Destructors (ICML 2018 · 8 citations)
- 2019 · Do Deep Generative Models Know What They Don't Know? (ICLR 2019 · 55 citations)
- 2020 · Log-Likelihood Ratio Minimizing Flows: Towards Robust and Quantifiable Neural Distribution Alignment (NeurIPS 2020 · 5 citations)
- 2022 · Efficient Maximal Coding Rate Reduction by Variational Forms (CVPR 2022 · 5 citations)
- 2023 · Positive Distribution Pollution: Rethinking Positive Unlabeled Learning from a Unified Perspective (AAAI 2023 · 2 citations)
- 2024 · An amortized approach to non-linear mixed-effects modeling based on neural posterior estimation (ICML 2024 · 4 citations)
- 2025 · Enhanced Importance Sampling Through Latent Space Exploration in Normalizing Flows (AAAI 2025 · 1 citation)
- 2026 · Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomics (ICML 2026 · 0 citations)
Earliest papers here
- Masked Autoregressive Flow for Density Estimation (NeurIPS 2017 · 201 citations)
- Deep Density Destructors (ICML 2018 · 8 citations)
- Do Deep Generative Models Know What They Don't Know? (ICLR 2019 · 55 citations)
What has been tried
25 papers here report 230 measured changes (a component added, removed or swapped, compared with a baseline): helped 59%, no clear effect 11%, hurt 26%, mixed 4%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Allow a full covariance matrix for random effects instead of a diagonal covariance matrix | Detailed single-cell mRNA transfection model · fit of population variance | Helped | An amortized approach to non-linear mixed-effects modeling based on neural posterior estimation (ICML 2024 · 4 citations) | |
| Replace learned density estimates with ground-truth density estimates for graph edge weights | Five two-dimensional datasets · Log Path Ratio | Helped | Learning Distances from Data with Normalizing Flows and Score Matching (ICML 2025 · 0 citations) | |
| Use LPCA-calibrated dataset-specific threshold instead of fixed τ | CIFAR10 vs. SVHN · AUC-ROC | 0.737 -> 0.936 | Helped | A Geometric Explanation of the Likelihood OOD Detection Paradox (ICML 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 8 papers of this approach from 2024–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- MNIST · 3 of 8 papers
- CIFAR-10 · 2 of 8 papers
- SVHN · 2 of 8 papers
- FMNIST · 2 of 8 papers
- CelebA · 2 of 8 papers
- TinyImageNet · 2 of 8 papers
- OMNIGLOT · 2 of 8 papers
Recent preprints
Loading recent arXiv preprints…
Use it
Download BibTeX of 26 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Normalizing flow training and design (88 papers)
- Expressive invertible architectures (76 papers)
Elsewhere in the topic
- Likelihood-free and graphical model estimation (161 papers)
- Flow matching and flow sampling (144 papers)
- Posterior sampling with diffusion guidance (137 papers)
- Score-based generative modeling analysis (128 papers)
- Structured latent VAEs (125 papers)
- Exact inference for probabilistic programs (103 papers)
- Disentangled and discrete latent models (92 papers)
- Latent variable marginalization and estimators (91 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).