Wenxuan Wang

Wenxuan Wang (王文轩)

I'm currently a Technical Staff at Alibaba Group, Token Foundry. I received my Ph.D. at the Institute of Automation, Chinese Academy of Sciences and the Beijing Academy of Artificial Intelligence, co-supervised by Prof. Jing Liu and Dr. Xinlong Wang. My research interests span Foundation Models, Native Multimodal Models, Generative Models.

Email / Google Scholar / Github / Curriculum Vitae

Education

University of Science and Technology Beijing, Sept. 2016 - Jun. 2020, B.S. in School of Automation.

University of Science and Technology Beijing, Sept. 2020 - Jun. 2023, M.S. in School of Automation.

Institute of Automation, Chinese Academy of Sciences, Sept. 2023 - Jun. 2026, Ph.D. in Zidongtaichu Foundation Model Research Center.

Beijing Academy of Artificial Intelligence, Sept. 2023 - Jun. 2026, Ph.D. in Multimodal Large Model Research Center.

Recent Projects

* indicates equal contribution

Emu3.5: Native Multimodal Models are World Learners
Yufeng Cui*, Honghao Chen*, Haoge Deng*, Xu Huang*, Xinghang Li*, Jirong Liu*, Yang Liu*, Zhuoyan Luo*, Jinsheng Wang*, Wenxuan Wang*, Yueze Wang*, Chengyuan Wang*, Fan Zhang*, Yingli Zhao*, Ting Pan, Xianduo Li, Zecheng Hao, Wenxuan Ma, Zhuo Chen, Yulong Ao, Tiejun Huang, Zhongyuan Wang, Xinlong Wang
arXiv, 2025
[Paper] [Page] [Code]

a large-scale multimodal world model that natively predicts the next state across vision and language

First-author Publications

* indicates equal contribution

	End-to-End Vision Tokenizer Tuning Wenxuan Wang, Fan Zhang, Yufeng Cui, Haiwen Diao, Zhuoyan Luo, Huchuan Lu, Jing Liu, Xinlong Wang NeurIPS, 2025 [Paper] an end-to-end vision tokenizer tuning approach that enables joint optimization between vision tokenization and target autoregressive tasks
	Towards Unified Referring Expression Segmentation Across Omni-Level Visual Target Granularities Jing Liu, Wenxuan Wang, Yisi Zhang, Yepeng Tang, Xingjian He, Longteng Guo, Tongtian Yue, Xinlong Wang arXiv, 2025 [Paper] takes a step further towards visual granularity unified RES task
	Image Difference Grounding with Natural Language Wenxuan Wang, Zijia Zhao, Yisi Zhang*, Yepeng Tang, Erdong Hu, Xinlong Wang, Jing Liu arXiv, 2025 [Paper] push towards precisely localizing visual differences based on user instructions
	Diffusion Feedback Helps CLIP See Better Wenxuan Wang, Quan Sun, Fan Zhang, Yepeng Tang, Jing Liu, Xinlong Wang ICLR, 2025 [Paper] [Page] [Code] leverages generative feedback from text-to-image diffusion models to optimize CLIP representations, with only images (without corresponding text)
	Beyond Literal Descriptions: Understanding and Locating Open-World Objects Aligned with Human Intentions Wenxuan Wang, Yisi Zhang, Xingjian He, Yichen Yan, Zijia Zhao, Xinlong Wang, Jing Liu ACL, 2024 (Findings) [Paper] takes a step further to the intention-driven visual-language understanding and promotes classic visual grounding towards human intention interpretation
	Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation Wenxuan Wang, Tongtian Yue, Yisi Zhang, Longteng Guo, Xingjian He, Xinlong Wang, Jing Liu CVPR, 2024 [Paper] [Page] [Code] takes a step further to finer-grained part-level referring expression segmentation task
	CM-MaskSD: Cross-Modality Masked Self-Distillation for Referring Image Segmentation Wenxuan Wang, Jing Liu, Xingjian He, Yisi Zhang, Chen Chen, Jiachen Shen, Yan Zhang, Jiangyun Li IEEE-TMM, 2024 [Paper] a new cross-modality masked self-distillation framework for referring image segmentation task

Co-author Publications

* indicates equal contribution

	EVEv2: Improved Baselines for Encoder-Free Vision-Language Models Haiwen Diao, Xiaotong Li, Yufeng Cui, Yueze Wang, Haoge Deng, Ting Pan, Wenxuan Wang, Huchuan Lu, Xinlong Wang ICCV, 2025 (highlight) [Paper] [Code] encoder-free vision-language models
	Unified Vision-Language-Action Model Yuqi Wang, Xinghang Li, Wenxuan Wang, Junbo Zhang, Yingyan Li, Yuntao Chen, Xinlong Wang, Zhaoxiang Zhang arXiv, 2025 [Paper] [Page] [Code] unified vision-language-action model for embodied intelligence

Honors and Awards

2025 National Scholarship (Ph.D.)

2022 National Scholarship (Master)

Website Template