UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encoding

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jiao, Yang, Qiu, Haibo, Jie, Zequn, Chen, Shaoxiang, Chen, Jingjing, Ma, Lin, Jiang, Yu-Gang
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908304252338176
author Jiao, Yang
Qiu, Haibo
Jie, Zequn
Chen, Shaoxiang
Chen, Jingjing
Ma, Lin
Jiang, Yu-Gang
author_facet Jiao, Yang
Qiu, Haibo
Jie, Zequn
Chen, Shaoxiang
Chen, Jingjing
Ma, Lin
Jiang, Yu-Gang
contents We introduce UniToken, an auto-regressive generation model that encodes visual inputs through a combination of discrete and continuous representations, enabling seamless integration of unified visual understanding and image generation tasks. Unlike previous approaches that rely on unilateral visual representations, our unified visual encoding framework captures both high-level semantics and low-level details, delivering multidimensional information that empowers heterogeneous tasks to selectively assimilate domain-specific knowledge based on their inherent characteristics. Through in-depth experiments, we uncover key principles for developing a unified model capable of both visual understanding and image generation. Extensive evaluations across a diverse range of prominent benchmarks demonstrate that UniToken achieves state-of-the-art performance, surpassing existing approaches. These results establish UniToken as a robust foundation for future research in this domain. The code and models are available at https://github.com/SxJyJay/UniToken.
format Preprint
id arxiv_https___arxiv_org_abs_2504_04423
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encoding
Jiao, Yang
Qiu, Haibo
Jie, Zequn
Chen, Shaoxiang
Chen, Jingjing
Ma, Lin
Jiang, Yu-Gang
Computer Vision and Pattern Recognition
Artificial Intelligence
We introduce UniToken, an auto-regressive generation model that encodes visual inputs through a combination of discrete and continuous representations, enabling seamless integration of unified visual understanding and image generation tasks. Unlike previous approaches that rely on unilateral visual representations, our unified visual encoding framework captures both high-level semantics and low-level details, delivering multidimensional information that empowers heterogeneous tasks to selectively assimilate domain-specific knowledge based on their inherent characteristics. Through in-depth experiments, we uncover key principles for developing a unified model capable of both visual understanding and image generation. Extensive evaluations across a diverse range of prominent benchmarks demonstrate that UniToken achieves state-of-the-art performance, surpassing existing approaches. These results establish UniToken as a robust foundation for future research in this domain. The code and models are available at https://github.com/SxJyJay/UniToken.
title UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encoding
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2504.04423