Ming-UniVision: Joint Image Understanding and Generation with a Unified Continuous Tokenizer
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915538681200640 |
|---|---|
| author | Huang, Ziyuan Zheng, DanDan Zou, Cheng Liu, Rui Wang, Xiaolong Ji, Kaixiang Chai, Weilong Sun, Jianxin Wang, Libin Lv, Yongjie Huang, Taozhi Liu, Jiajia Guo, Qingpei Yang, Ming Chen, Jingdong Zhou, Jun |
| author_facet | Huang, Ziyuan Zheng, DanDan Zou, Cheng Liu, Rui Wang, Xiaolong Ji, Kaixiang Chai, Weilong Sun, Jianxin Wang, Libin Lv, Yongjie Huang, Taozhi Liu, Jiajia Guo, Qingpei Yang, Ming Chen, Jingdong Zhou, Jun |
| contents | Visual tokenization remains a core challenge in unifying visual understanding and generation within the autoregressive paradigm. Existing methods typically employ tokenizers in discrete latent spaces to align with the tokens from large language models, where the quantization errors can limit semantic expressiveness and degrade the capability of vision-language understanding. To address this, we introduce MingTok, a new family of visual tokenizers with a continuous latent space, for unified autoregressive generation and understanding. While understanding tasks favor discriminative high-dimensional features, generation tasks prefer compact low-level codes. Thus, to reconcile these competing demands, MingTok adopts a three-stage sequential architecture involving low-level encoding, semantic expansion, and visual reconstruction. Built on top of it, Ming-UniVision eliminates the need for task-specific visual representations, and unifies diverse vision-language tasks under a single autoregrsssive prediction paradigm. By formulating both understanding and generation as next-token prediction in a shared continuous space, it seamlessly supports multi-round, in-context tasks such as iterative understanding, generation and editing. Empirically, we find that using a unified continuous visual representation reconciles the competing requirements on the tokenizers by the understanding and generation tasks, thereby leading to state-of-the-art level performance across both domains. We hope our findings will facilitate unified visual tokenization in the continuous domain. Inference code and model weights are released to benefit community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_06590 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Ming-UniVision: Joint Image Understanding and Generation with a Unified Continuous Tokenizer Huang, Ziyuan Zheng, DanDan Zou, Cheng Liu, Rui Wang, Xiaolong Ji, Kaixiang Chai, Weilong Sun, Jianxin Wang, Libin Lv, Yongjie Huang, Taozhi Liu, Jiajia Guo, Qingpei Yang, Ming Chen, Jingdong Zhou, Jun Computer Vision and Pattern Recognition Visual tokenization remains a core challenge in unifying visual understanding and generation within the autoregressive paradigm. Existing methods typically employ tokenizers in discrete latent spaces to align with the tokens from large language models, where the quantization errors can limit semantic expressiveness and degrade the capability of vision-language understanding. To address this, we introduce MingTok, a new family of visual tokenizers with a continuous latent space, for unified autoregressive generation and understanding. While understanding tasks favor discriminative high-dimensional features, generation tasks prefer compact low-level codes. Thus, to reconcile these competing demands, MingTok adopts a three-stage sequential architecture involving low-level encoding, semantic expansion, and visual reconstruction. Built on top of it, Ming-UniVision eliminates the need for task-specific visual representations, and unifies diverse vision-language tasks under a single autoregrsssive prediction paradigm. By formulating both understanding and generation as next-token prediction in a shared continuous space, it seamlessly supports multi-round, in-context tasks such as iterative understanding, generation and editing. Empirically, we find that using a unified continuous visual representation reconciles the competing requirements on the tokenizers by the understanding and generation tasks, thereby leading to state-of-the-art level performance across both domains. We hope our findings will facilitate unified visual tokenization in the continuous domain. Inference code and model weights are released to benefit community. |
| title | Ming-UniVision: Joint Image Understanding and Generation with a Unified Continuous Tokenizer |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.06590 |