Research map / Probabilistic and scientific modeling / Variational and generative sampling
Likelihood-free and graphical model estimation
Estimates intractable models using kernel ABC, Stein tests and sparse graphical models.
161 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Variational and generative sampling, keyword cluster “estimation · probabilistic · variables”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
43 papers in 2023–24 and 27 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 10.4% to 5.4%.
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
- 2016 · A Kernelized Stein Discrepancy for Goodness-of-fit Tests (ICML 2016 · 154 citations)
- 2017 · Density estimation using Real NVP (ICLR 2017 · 735 citations)
- 2018 · Gaussian Process Conditional Density Estimation (NeurIPS 2018 · 30 citations)
- 2019 · Hierarchical Discrete Distribution Decomposition for Match Density Estimation (CVPR 2019 · 277 citations)
- 2020 · Modeling Noisy Annotations for Crowd Counting (NeurIPS 2020 · 51 citations)
- 2021 · Sliced Kernelized Stein Discrepancy (ICLR 2021 · 14 citations)
- 2022 · Unsupervised Anomaly Detection by Robust Density Estimation (AAAI 2022 · 27 citations)
- 2023 · Spearman Rank Correlation Screening for Ultrahigh-Dimensional Censored Data (AAAI 2023 · 16 citations)
- 2024 · Effectiveness of Constant Stepsize in Markovian LSA and Statistical Inference (AAAI 2024 · 3 citations)
- 2025 · Conditional Diffusion Models Based Conditional Independence Testing (AAAI 2025 · 1 citation)
- 2026 · Relative Entropy Estimation in Function Space: Theory and Applications to Trajectory Inference (ICML 2026 · 0 citations)
Earliest papers here
- A Kernelized Stein Discrepancy for Goodness-of-fit Tests (ICML 2016 · 154 citations)
- Markov-modulated Marked Poisson Processes for Check-in Data (ICML 2016 · 16 citations)
- Consistent Estimation of Functions of Data Missing Non-Monotonically and Not at Random (NeurIPS 2016 · 11 citations)
What has been tried
138 papers here report 1,139 measured changes (a component added, removed or swapped, compared with a baseline): helped 67%, no clear effect 11%, hurt 17%, mixed 6%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Use response-distribution transformation rather than h(Y) = Y | Outlier-contaminated simulations · empirical rejection rate (power) | Helped | Robust group and simultaneous inferences for high-dimensional single index model (NeurIPS 2024 · 1 citation) | |
| Use two-step GMM instead of one-step GMM (identity weighting matrix) | Latent covariance estimation · Asymptotic variance | Helped | A Sample Efficient Conditional Independence Test in the Presence of Discretization (ICML 2025 · 0 citations) | |
| Use a diffusion-model estimated score instead of the oracle score | 2D spiral · MISE | Mixed | SD-KDE: Score-Debiased Kernel Density Estimation (NeurIPS 2025 · 0 citations) | |
| Increase time discretization from N=10 to N=50 | Action convergence with time discretization · action | 66.06 -> 30.10 | Helped | PDPO: Parametric Density Path Optimization (NeurIPS 2025 · 0 citations) |
| Use diagonal or K-FAC Hessian approximation instead of the full Hessian | CIFAR-10 · Spearman's R | R>0.9 | No clear effect | Laplace Sample Information: Data Informativeness Through a Bayesian Lens (ICLR 2025 · 0 citations) |
| Use features from a deeper layer rather than shallower layers | CIFAR-10 · OOD detection performance | Helped | ConjNorm: Tractable Density Estimation for Out-of-Distribution Detection (ICLR 2024 · 0 citations) | |
| Add more informative constraints on race and income | Synthetic conditional mean · bound width | Helped | Statistical Inference Under Constrained Selection Bias (ICML 2024 · 0 citations) | |
| use the SDO kernel instead of Gaussian or Laplace kernels | ADBench anomaly detection · AUC-ROC | Helped | Sobolev Space Regularised Pre Density Models (ICML 2024 · 0 citations) | |
| Use the exponential integrator instead of the NAG-like integrator for the parameter component | Toy hierarchical model · optimization performance | Helped | Momentum Particle Maximum Likelihood (ICML 2024 · 0 citations) | |
| add entropy regularization | TM · Entropy | 1.0 (0.198) -> 1.26 (0.022) | Helped | Sourcerer: Sample-based Maximum Entropy Source Distribution Estimation (NeurIPS 2024 · 0 citations) |
| Add control variates for power enhancement | Simulation, β = 1.4 · power | 0.44 -> 0.8 | Helped | Covariate Shift Corrected Conditional Randomization Test (NeurIPS 2024 · 0 citations) |
| Add hard truncation of low non-corruption probabilities after overfitting begins | CIFAR10, pairflip 45% noise · test accuracy | Helped | Adaptive Robust Learning using Latent Bernoulli Variables (ICML 2024) |
What reviewers will expect
Datasets and baselines named in the 18 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Baselines
- DeepHit · 2 of 18 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 70 papers; from OpenAlex and the papers' affiliations).
- Carnegie Mellon University · 5 papers
- East China Normal University · 3 papers
- Cornell University · 3 papers
- Peking University · 3 papers
- University College London · 3 papers
- Amazon · 3 papers
- University of Texas, Austin · 3 papers
- New York University Shanghai · 2 papers
- New York University · 2 papers
- Boston University · 2 papers
Use it
Download BibTeX of 40 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
- Tractable probabilistic models (90 papers)
- Exact inference for probabilistic programs (103 papers)
- Latent variable marginalization and estimators (91 papers)
- Gradient estimation for discrete variables (74 papers)
- Mutual information estimation (27 papers)
Elsewhere in the topic
- 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)
- Disentangled and discrete latent models (92 papers)
- Variational inference theory and gradients (89 papers)
- Hamiltonian Monte Carlo (89 papers)
- Regularizing VAE latent information (89 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).