cs.CVNov 9, 2026

dNaViT: Discrete Codes are General-Purpose Visual Representations

Authors: Xiaotong Li, Zhengrong Yue, Xu Huang, Jialv Zou, Minhao Jing, Wenhao Yan, Haodong Li, Hai Huang, 

Abstract

Discrete autoregressive modeling, notably through next-token prediction, has become a cornerstone of modern language models. Extending this paradigm to vision allows visual and language models to converge on a unified framework, thereby fostering shared methodology and infrastructure. This work investigates a central question: how far can a finite discrete vocabulary go in representing the richness of the visual world? We present dNaViT, a Discrete Native-resolution Vision Transformer that produces discrete visual primitives analogous to language tokens. Built upon a frozen pretrained encoder, dNaViT applies lightweight residual modulation with a dual-branch objective to inject reconstruction-aware cues into semantic features. By leveraging grouped product quantization, it scales discrete capacity without inflating spatial sequence lengths, while a balanced training recipe maintains both semantic and pixel-level information. Consequently, dNaViT functions as a versatile discrete visual tokenizer, enabling unified encoding across downstream tasks alongside high-fidelity image decoding. Extensive evaluations show that dNaViT matches or exceeds task-specific baselines across 12 visual understanding benchmarks, 4 generation benchmarks, and an image editing benchmark. These findings demonstrate that a compact discrete vocabulary can serve as a universal visual currency and unlock the full potential of fully unified, native multimodal methodology.

Figures & tables

Explore similar work

CardsList
  1. Can Discrete Tokens Serve as Semantically Lossless Representations of Visual Signals?

    Oct 27, 2026Jialv Zou, Xiaotong Li, Zhengrong Yue +6Discrete Visual TokenizationVisual Representation Learning

  2. A More Word-like Image Tokenization for MLLMs

    May 18, 2026Hyun Lee, Hyemin Jeong, Yejin Kim +4Multimodal UnderstandingEfficient VLM Inference