mlsanityData to 14 Oct 2026

Preventing posterior collapse in variational autoencoders for text generation via decoder regularization

arXiv 2110.14945 ↗First posted 28 Oct 2021cs.LG cs.CLcited by 2

Not on the map. Why?

Abstract (arXiv)

Variational autoencoders trained to minimize the reconstruction error are sensitive to the posterior collapse problem, that is the proposal posterior distribution is always equal to the prior. We propose a novel regularization method based on fraternal dropout to prevent posterior collapse. We evaluate our approach using several metrics and observe improvements in all the tested configurations.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 19 of its 23 references that are in the map. First posted on arXiv 28 Oct 2021.

Builds on

19 of 23 references.

References · 19

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015A large annotated corpus for learning natural language inferencecited by 1,847
  3. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  4. 2015Generating Sentences from a Continuous Spacecited by 612
  5. 2015Importance Weighted Autoencoderscited by 456
  6. 2016Deep Biaffine Attention for Neural Dependency Parsingcited by 334
  7. 2017Style Transfer from Non-Parallel Text by Cross-Alignmentcited by 248
  8. 2017Regularizing and Optimizing LSTM Language Modelscited by 205
  9. 2015A Theoretically Grounded Application of Dropout in Recurrent Neural Networkscited by 161
  10. 2019Lagging Inference Networks and Posterior Collapse in Variational Autoencoderscited by 124
  11. 2017Improved Variational Autoencoders for Text Modeling using Dilated Convolutionscited by 81
  12. 2016Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activationscited by 54
7 more
  1. 2018Avoiding Latent Variable Collapse With Generative Skip Modelscited by 47
  2. 2019A Surprisingly Effective Fix for Deep Latent Variable Modeling of Textcited by 32
  3. 2017Fraternal Dropoutcited by 14
  4. 2020Preventing Posterior Collapse with Levenshtein Variational Autoencodercited by 12
  5. 2019Effective Estimation of Deep Generative Language Modelscited by 9
  6. 2018h-detach: Modifying the LSTM Gradient Towards Better Optimizationcited by 5
  7. 2020SentenceMIM: A Latent Variable Language Modelcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 0; 2024: 0; 2025: 0; 2026: 1.

2 papers.

  1. 2022Improving Variational Autoencoders with Density Gap-based Regularizationcited by 3
  2. 2026Entropic Auto-Encoding via Implicit Free-Energy Minimizationcited by 0