# VAEVQ: Enhancing Discrete Visual Tokenization through Variational Modeling Source: [VAEVQ: Enhancing Discrete Visual Tokenization through Variational Modeling]() Collection index: [llms.txt]() Data updated: 2025-11-12T05:05:45.634Z Authors: Sicheng Yang, Xing Hu, Qiang Wu, Dawei Yang Published: 2025-11-10T09:07:23.000Z ## Abstract Vector quantization (VQ) transforms continuous image features into discrete representations, providing compressed, tokenized inputs for generative models. However, VQ-based frameworks suffer from several issues, such as non-smooth latent spaces, weak alignment between representations before and after quantization, and poor coherence between the continuous and discrete domains. These issues lead to unstable codeword learning and underutilized codebooks, ultimately degrading the performance of both reconstruction and downstream generation tasks. To this end, we propose VAEVQ, which comprises three key components: (1) Variational Latent Quantization (VLQ), replacing the AE with a VAE for quantization to leverage its structured and smooth latent space, thereby facilitating more effective codeword activation; (2) Representation Coherence Strategy (RCS), adaptively modulating the alignment strength between pre- and post-quantization features to enhance consistency and prevent overfitting to noise; and (3) Distribution Consistency Regularization (DCR), aligning the entire codebook distribution with the continuous latent distribution to improve utilization. Extensive experiments on two benchmark datasets demonstrate that VAEVQ outperforms state-of-the-art methods. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Reconstruction on ImageNet-1k 256x256](): Image Reconstruction; ImageNet-1K 256x256; rFID: 1.14. ## Original sources and implementation [Original paper]() [PDF]()