Research map / Probabilistic and scientific modeling / Variational and generative sampling
Hierarchical VAEs and posterior collapse
Deepens latent hierarchies and prevents posterior collapse in autoencoder models.
59 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in Variational and generative sampling, keyword cluster “vae · autoencoders · latent”
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.
11 papers in 2023–24 and 16 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 2.7% to 3.2%.
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 · Grammar Variational Autoencoder (ICML 2017 · 112 citations)
- 2018 · Latent Space Oddity: on the Curvature of Deep Generative Models (ICLR 2018 · 24 citations)
- 2019 · BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling (NeurIPS 2019 · 116 citations)
- 2020 · Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder (NeurIPS 2020 · 71 citations)
- 2021 · Property Controllable Variational Autoencoder via Invertible Mutual Dependence (ICLR 2021 · 9 citations)
- 2022 · FactorVAE: A Probabilistic Dynamic Factor Model Based on Variational Autoencoder for Predicting Cross-Sectional Stock Returns (AAAI 2022 · 73 citations)
- 2023 · Latent Autoregressive Source Separation (AAAI 2023 · 6 citations)
- 2024 · Learned Lossless Image Compression based on Bit Plane Slicing (CVPR 2024 · 11 citations)
- 2025 · Robust Multimodal Survival Prediction with Conditional Latent Differentiation Variational AutoEncoder (CVPR 2025 · 13 citations)
- 2026 · Geodesic Calculus on Implicitly Defined Latent Manifolds (ICML 2026 · 0 citations)
Earliest papers here
- Grammar Variational Autoencoder (ICML 2017 · 112 citations)
- A Diversified Generative Latent Variable Model for WiFi-SLAM (AAAI 2017 · 8 citations)
- Maximum Reconstruction Estimation for Generative Latent-Variable Models (AAAI 2017 · 0 citations)
What has been tried
43 papers here report 448 measured changes (a component added, removed or swapped, compared with a baseline): helped 70%, no clear effect 9%, hurt 17%, 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 |
|---|---|---|---|---|
| Use VIB-Trans instead of the TransMIL encoder for WSI representation | Overall, complete modality · C-index | 0.701 -> 0.719 | Helped | Robust Multimodal Survival Prediction with Conditional Latent Differentiation Variational AutoEncoder (CVPR 2025 · 13 citations) |
| add plane autoregression | CIFAR10 · theoretical BPD | 3.458 -> 3.052 | Helped | Learned Lossless Image Compression based on Bit Plane Slicing (CVPR 2024 · 11 citations) |
| set flattening strength to λ=10 instead of λ=0 | Synthetic negative binomial · 3-NN overlap | 0.66 -> 0.80 | Helped | Enforcing Latent Euclidean Geometry in Single-Cell VAEs for Manifold Interpolation (ICML 2025 · 0 citations) |
| Set RBF width σ to 0.05 rather than 1.0 at λ=1.0 | AnimeFace 32×32 · CMMD | 0.56 -> 0.74 | Hurt | Towards Scalable Topological Regularizers (ICLR 2025 · 0 citations) |
| Learn projection variance σ² instead of fixing it | MNIST · KNN accuracy | Helped | Preventing Model Collapse in Gaussian Process Latent Variable Models (ICML 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 11 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- MNIST · 3 of 11 papers
- Fashion-MNIST · 3 of 11 papers
Baselines
- VAE · 2 of 11 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 27 papers; from OpenAlex and the papers' affiliations).
- Helmholtz Zentrum München · 2 papers
Use it
Download BibTeX of 38 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
- Structured latent VAEs (125 papers)
- Regularizing VAE latent information (89 papers)
- Disentangled and discrete latent models (92 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)
- Exact inference for probabilistic programs (103 papers)
- Latent variable marginalization and estimators (91 papers)
- Tractable probabilistic models (90 papers)
- Variational inference theory and gradients (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).