Toward Architecture-Agnostic Local Control of Posterior Collapse in VAEs

• URL: http://arxiv.org/abs/2508.12530v1
• Date: Sun, 17 Aug 2025 23:45:41 GMT
• Title: Toward Architecture-Agnostic Local Control of Posterior Collapse in VAEs
• Abstract summary: Variational autoencoders (VAEs) are known to suffer from posterior collapse, a phenomenon that reduces the diversity of generated samples.<n>We propose Latent Reconstruction(LR) loss, inspired by mathematical properties of injective and composite functions, to control posterior collapse without restriction to a specific architecture.<n>We experimentally evaluate our approach, which controls posterior collapse on varied datasets such as MNIST, fashionMNIST, Omniglot, CelebA, and FFHQ.
• Score: 0.6144680854063939
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Variational autoencoders (VAEs), one of the most widely used generative models, are known to suffer from posterior collapse, a phenomenon that reduces the diversity of generated samples. To avoid posterior collapse, many prior works have tried to control the influence of regularization loss. However, the trade-off between reconstruction and regularization is not satisfactory. For this reason, several methods have been proposed to guarantee latent identifiability, which is the key to avoiding posterior collapse. However, they require structural constraints on the network architecture. For further clarification, we define local posterior collapse to reflect the importance of individual sample points in the data space and to relax the network constraint. Then, we propose Latent Reconstruction(LR) loss, which is inspired by mathematical properties of injective and composite functions, to control posterior collapse without restriction to a specific architecture. We experimentally evaluate our approach, which controls posterior collapse on varied datasets such as MNIST, fashionMNIST, Omniglot, CelebA, and FFHQ.

Related papers

• Prototype-Anchored Generalized Manifold Regression for Unknown-Domain Object Detection [58.25418970608328]
  We study Single-Domain Generalized Object Detection (Single-DGOD), which aims to transfer a detector trained on a single source domain to multiple unseen domains.<n>Existing methods mainly rely on simulation-driven strategies, such as data augmentation or textual prompts, to enlarge the training distribution.<n>We propose Manifold Regression with Visual-Text Dual Chain-of-Thought (MR-DCoT), which formulates unknown-domain generalization as a manifold regression problem.
  arXiv  Detail & Related papers  (2026-07-08T09:27:15Z)
• MaRS: Robust Out-of-Distribution Detection via Mahalanobis Residual Scoring [2.0480448867367476]
  We introduce MaRS, a label-free OOD detector that learns an in-distribution manifold using a lightweight autoencoder and measures deviation via a Mahalanobis distance on reconstruction residuals.<n>MaRS outperforms established confidence-, distance-, and reconstruction-based baselines, while remaining fully post-hoc and lightweight.
  arXiv  Detail & Related papers  (2026-06-21T19:45:41Z)
• Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction [52.32112533846212]
  MRI reconstruction is an inherently ill-posed inverse problem, since incomplete measurements admit many plausible solutions.<n>We address this limitation by moving reconstruction to discrete multi-scale latent space and posing it as autoregressive next-acceleration-scale prediction.<n>We show that our approach delivers improved reconstruction performance across diverse sampling patterns under extreme undersampling.
  arXiv  Detail & Related papers  (2026-05-19T04:40:50Z)
• Historical Consensus: Preventing Posterior Collapse via Iterative Selection of Gaussian Mixture Priors [4.284407845277412]
  Variational autoencoders (VAEs) frequently suffer from posterior collapse.<n>We propose a fundamentally different approach: instead of avoiding collapse through architectural constraints, we eliminate the possibility of collapse altogether.
  arXiv  Detail & Related papers  (2026-03-11T16:19:07Z)
• How Far Are We from Generating Missing Modalities with Foundation Models? [49.425856207329524]
  We propose an agentic framework tailored for missing modality reconstruction.<n>Our method reduces FID for missing image reconstruction by at least 14% and MER for missing text reconstruction by at least 10% compared to baselines.
  arXiv  Detail & Related papers  (2025-06-04T03:22:44Z)
• Verifiably Forgotten? Gradient Differences Still Enable Data Reconstruction in Federated Unlearning [12.13642771298657]
  Federated Unlearning (FU) has emerged as a critical compliance mechanism for data privacy regulations.<n>We propose Inverting Gradient difference to Forgotten data (IGF), a novel learning-based reconstruction attack framework.<n>IGF incorporates a tailored pixel-level inversion model optimized via a composite loss that captures both structural and semantic cues.
  arXiv  Detail & Related papers  (2025-05-16T10:28:30Z)
• DiAD: A Diffusion-based Framework for Multi-class Anomaly Detection [55.48770333927732]
  We propose a Difusion-based Anomaly Detection (DiAD) framework for multi-class anomaly detection. It consists of a pixel-space autoencoder, a latent-space Semantic-Guided (SG) network with a connection to the stable diffusion's denoising network, and a feature-space pre-trained feature extractor. Experiments on MVTec-AD and VisA datasets demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2023-12-11T18:38:28Z)
• Subspace-Guided Feature Reconstruction for Unsupervised Anomaly Localization [5.085309164633571]
  Unsupervised anomaly localization plays a critical role in industrial manufacturing. Most recent methods perform feature matching or reconstruction for the target sample with pre-trained deep neural networks. We propose a novel subspace-guided feature reconstruction framework to pursue adaptive feature approximation for anomaly localization.
  arXiv  Detail & Related papers  (2023-09-25T06:58:57Z)
• Neural Network Training Strategy to Enhance Anomaly Detection Performance: A Perspective on Reconstruction Loss Amplification [2.7503452586560484]
  Unsupervised anomaly detection (UAD) is a widely adopted approach in industry due to rare anomaly occurrences and data imbalance. Recent studies have pursued to contain the generalization capability of their UAD models in reconstruction from different perspectives. We propose a loss landscape sharpening method by amplifying the reconstruction loss, dubbed Loss AMPlification.
  arXiv  Detail & Related papers  (2023-08-28T14:06:36Z)
• Controlling Posterior Collapse by an Inverse Lipschitz Constraint on the Decoder Network [37.43129242269175]
  Variational autoencoders (VAEs) are one of the deep generative models that have experienced enormous success over the past decades. VAEs suffer from a problem called posterior collapse, which occurs when the encoder coincides, or collapses. We introduce an inverse Lipschitz neural network into the decoder and, based on this architecture, provide a new method that can control in a simple and clear manner the degree of posterior collapse.
  arXiv  Detail & Related papers  (2023-04-25T12:45:06Z)
• Mind the Backbone: Minimizing Backbone Distortion for Robust Object Detection [52.355018626115346]
  Building object detectors that are robust to domain shifts is critical for real-world applications. We propose to use Relative Gradient Norm as a way to measure the vulnerability of a backbone to feature distortion. We present recipes to boost OOD robustness for both types of backbones.
  arXiv  Detail & Related papers  (2023-03-26T14:50:43Z)
• Discouraging posterior collapse in hierarchical Variational Autoencoders using context [19.47169312443202]
  There is a consensus that the top-down hierarchical VAEs allow effective learning of deep latent structures and avoid problems like posterior collapse. Here, we show that this is not necessarily the case, and the problem of collapsing posteriors remains. We propose a deep hierarchical VAE with a context on top. Specifically, we use a Discrete Cosine Transform to obtain the last latent variable.
  arXiv  Detail & Related papers  (2023-02-20T13:44:47Z)
• Making Reconstruction-based Method Great Again for Video Anomaly Detection [64.19326819088563]
  Anomaly detection in videos is a significant yet challenging problem. Existing reconstruction-based methods rely on old-fashioned convolutional autoencoders. We propose a new autoencoder model for enhanced consecutive frame reconstruction.
  arXiv  Detail & Related papers  (2023-01-28T01:57:57Z)
• Posterior Collapse and Latent Variable Non-identifiability [54.842098835445]
  We propose a class of latent-identifiable variational autoencoders, deep generative models which enforce identifiability without sacrificing flexibility. Across synthetic and real datasets, latent-identifiable variational autoencoders outperform existing methods in mitigating posterior collapse and providing meaningful representations of the data.
  arXiv  Detail & Related papers  (2023-01-02T06:16:56Z)
• Self-Supervised Training with Autoencoders for Visual Anomaly Detection [61.62861063776813]
  We focus on a specific use case in anomaly detection where the distribution of normal samples is supported by a lower-dimensional manifold. We adapt a self-supervised learning regime that exploits discriminative information during training but focuses on the submanifold of normal examples. We achieve a new state-of-the-art result on the MVTec AD dataset -- a challenging benchmark for visual anomaly detection in the manufacturing domain.
  arXiv  Detail & Related papers  (2022-06-23T14:16:30Z)
• Simple and Effective Prevention of Mode Collapse in Deep One-Class Classification [93.2334223970488]
  We propose two regularizers to prevent hypersphere collapse in deep SVDD. The first regularizer is based on injecting random noise via the standard cross-entropy loss. The second regularizer penalizes the minibatch variance when it becomes too small.
  arXiv  Detail & Related papers  (2020-01-24T03:44:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.