Is Hierarchical Quantization Essential for Optimal Reconstruction?

• URL: http://arxiv.org/abs/2601.22244v1
• Date: Thu, 29 Jan 2026 19:09:50 GMT
• Title: Is Hierarchical Quantization Essential for Optimal Reconstruction?
• Authors: Shirin Reyhanian, Laurenz Wiskott,
• Abstract summary: We show that inadequate codebook utilization limits single-level VQ-VAEs and that overly high-dimensional embeddings destabilize quantization and increase codebook collapse.<n>Our results demonstrate that when representational budgets are matched, and codebook collapse is mitigated, single-level VQ-VAEs can match the reconstruction fidelity of hierarchical variants.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vector-quantized variational autoencoders (VQ-VAEs) are central to models that rely on high reconstruction fidelity, from neural compression to generative pipelines. Hierarchical extensions, such as VQ-VAE2, are often credited with superior reconstruction performance because they split global and local features across multiple levels. However, since higher levels derive all their information from lower levels, they should not carry additional reconstructive content beyond what the lower-level already encodes. Combined with recent advances in training objectives and quantization mechanisms, this leads us to ask whether a single-level VQ-VAE, with matched representational budget and no codebook collapse, can equal the reconstruction fidelity of its hierarchical counterpart. Although the multi-scale structure of hierarchical models may improve perceptual quality in downstream tasks, the effect of hierarchy on reconstruction accuracy, isolated from codebook utilization and overall representational capacity, remains empirically underexamined. We revisit this question by comparing a two-level VQ-VAE and a capacity-matched single-level model on high-resolution ImageNet images. Consistent with prior observations, we confirm that inadequate codebook utilization limits single-level VQ-VAEs and that overly high-dimensional embeddings destabilize quantization and increase codebook collapse. We show that lightweight interventions such as initialization from data, periodic reset of inactive codebook vectors, and systematic tuning of codebook hyperparameters significantly reduce collapse. Our results demonstrate that when representational budgets are matched, and codebook collapse is mitigated, single-level VQ-VAEs can match the reconstruction fidelity of hierarchical variants, challenging the assumption that hierarchical quantization is inherently superior for high-quality reconstructions.

Related papers

• Understanding Degradation with Vision Language Model [56.09241449206817]
  Understanding visual degradations is a critical yet challenging problem in computer vision.<n>We introduce DU-VLM, a multimodal chain-of-thought model trained with supervised fine-tuning and reinforcement learning.<n>We also introduce textbfDU-110k, a large-scale dataset comprising 110,000 clean-degraded pairs with grounded physical annotations.
  arXiv  Detail & Related papers  (2026-02-04T13:51:15Z)
• Generalized Radius and Integrated Codebook Transforms for Differentiable Vector Quantization [11.898954874548073]
  We introduce a unified surrogate framework that keeps hard assignments in the forward pass while making VQ fully differentiable.<n>GRIT-VQ consistently improves reconstruction error, generative quality, and accuracy compared to existing VQ variants.
  arXiv  Detail & Related papers  (2026-02-01T10:22:35Z)
• VQRAE: Representation Quantization Autoencoders for Multimodal Understanding, Generation and Reconstruction [83.50898344094153]
  VQRAE produces Continuous semantic features for image understanding and tokens for visual generation within a unified tokenizer.<n>Design enables negligible semantic information for maintaining the ability of multimodal understanding, discrete tokens.<n>VQRAE presents competitive performance on several benchmarks of visual understanding, generation and reconstruction.
  arXiv  Detail & Related papers  (2025-11-28T17:26:34Z)
• Scalable Training for Vector-Quantized Networks with 100% Codebook Utilization [60.294965457786844]
  Vector quantization (VQ) is a key component in discrete tokenizers for image generation.<n>VQBridge is a robust, scalable, and efficient projector based on the map function method.<n>FVQ attains 100% codebook usage even with a 262k-codebook.
  arXiv  Detail & Related papers  (2025-09-12T11:08:21Z)
• Hyperbolic Residual Quantization: Discrete Representations for Data with Latent Hierarchies [48.72319569157807]
  Residual Quantization (RQ) is widely used to generate discrete, multitoken representations for hierarchical data.<n>We propose Hyperbolic Residual Quantization (HRQ), which embeds data in a hyperbolic manifold.<n>HRQ imparts an inductive bias that aligns naturally with hierarchical branching.
  arXiv  Detail & Related papers  (2025-05-18T13:14:07Z)
• IQPFR: An Image Quality Prior for Blind Face Restoration and Beyond [56.99331967165238]
  Blind Face Restoration (BFR) addresses the challenge of reconstructing degraded low-quality (LQ) facial images into high-quality (HQ) outputs.<n>We propose a novel framework that incorporates an Image Quality Prior (IQP) derived from No-Reference Image Quality Assessment (NR-IQA) models.<n>Our method outperforms state-of-the-art techniques across multiple benchmarks.
  arXiv  Detail & Related papers  (2025-03-12T11:39:51Z)
• HyperVQ: MLR-based Vector Quantization in Hyperbolic Space [56.4245885674567]
  A common solution is to employ Vector Quantization (VQ) within VQ Variational Autoencoders (VQVAEs)<n>We introduce HyperVQ, a novel approach that formulates VQ as a hyperbolic Multinomial Logistic Regression (MLR) problem.<n>Our experiments demonstrate that HyperVQ matches traditional VQ in generative and reconstruction tasks, while surpassing it in discriminative performance.
  arXiv  Detail & Related papers  (2024-03-18T03:17:08Z)
• HQ-VAE: Hierarchical Discrete Representation Learning with Variational Bayes [18.57499609338579]
  We propose a novel framework to learn hierarchical discrete representation on the basis of the variational Bayes framework, called hierarchically quantized variational autoencoder (HQ-VAE) HQ-VAE naturally generalizes the hierarchical variants of VQ-VAE, such as VQ-VAE-2 and residual-quantized VAE (RQ-VAE) Our comprehensive experiments on image datasets show that HQ-VAE enhances codebook usage and improves reconstruction performance.
  arXiv  Detail & Related papers  (2023-12-31T01:39:38Z)
• Hierarchical Residual Learning Based Vector Quantized Variational Autoencoder for Image Reconstruction and Generation [19.92324010429006]
  We propose a multi-layer variational autoencoder method, we call HR-VQVAE, that learns hierarchical discrete representations of the data. We evaluate our method on the tasks of image reconstruction and generation.
  arXiv  Detail & Related papers  (2022-08-09T06:04:25Z)
• Dual-constrained Deep Semi-Supervised Coupled Factorization Network with Enriched Prior [80.5637175255349]
  We propose a new enriched prior based Dual-constrained Deep Semi-Supervised Coupled Factorization Network, called DS2CF-Net. To ex-tract hidden deep features, DS2CF-Net is modeled as a deep-structure and geometrical structure-constrained neural network. Our network can obtain state-of-the-art performance for representation learning and clustering.
  arXiv  Detail & Related papers  (2020-09-08T13:10:21Z)
• Hierarchical Quantized Autoencoders [3.9146761527401432]
  We motivate the use of a hierarchy of Vector Quantized Variencoders (VQ-VAEs) to attain high factors of compression. We show that a combination of quantization and hierarchical latent structure aids likelihood-based image compression. Our resulting scheme produces a Markovian series of latent variables that reconstruct images of high-perceptual quality.
  arXiv  Detail & Related papers  (2020-02-19T11:26:34Z)

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.