|
Junke Wang 「王君可」
I'm a research scientist at ByteDance Seed (San Jose, CA, USA). Before this, I obtained my BS and PhD degrees from Fudan University (Shanghai, China) in 2021 and 2026, supervised by Prof. Zuxuan Wu and Prof. Yu-Gang Jiang. During my PhD, I interned at multiple frontier AI labs, including Bytedance Seed, Meta FAIR, and Microsoft Cloud AI.
My research interest lies in multimodal AI and robotics. Currently, I work on multimodal generation, world models, and world action models. Feel free to reach out if you are looking for collaboration opportunities.
Email: jkwang0724 [at] gmail [dot] com
Google Scholar [Full publications] /
Github
|
|
Technical Reports
|
ARM: An AutoRegressive Large Multimodal Model with Unified Discrete Representations.
[Page]
Junke Wang, Xiao Wang, Jiacheng Pan, Xuefeng Hu, Feng Li, Jingxiang Sun, Chaorui Deng, ...
Weilin Huang, Zuxuan Wu, Haoqi Fan, Yu-Gang Jiang, Zhenheng Yang.
Bytedance Seed
|
SimpleAR: Pushing the Frontier of Autoregressive Visual Generation through Pretraining, SFT, and RL.
[Code]
Junke Wang, Zhi Tian, Xun Wang, Xinyu Zhang, Weilin Huang, Zuxuan Wu, Yu-Gang Jiang.
Bytedance Seed
|
Lingbot-VA 2.0: Native Video-Action Pretraining for Generalizable Robot Control.
[Page]
Core contributor for pretraining.
RobbyAnt
|
Perception Encoder: The best visual embeddings are not at the output of the network.
[Code]
[Page]
Core contributor for video pretraining and evaluation.
Meta FAIR
|
|
Publications
|
|
* denotes equal contribution, † denotes project lead.
|
DeRA: Decoupled Representation Alignment for Video Tokenization.
Pengbo Guo, Junke Wang†, Zhen Xing, Chengxu Liu, Daoguo Dong, Xueming Qian, Zuxuan Wu.
European Conference on Computer Vision (ECCV), 2026.
|
VideoLoom: A Video Large Language Model for Joint Spatial-Temporal Understanding.
[Code]
Jiapeng Shi, Junke Wang†, Zuyao You, Bo He, Zuxuan Wu.
International Conferenece on Machine Learning (ICML), 2026.
|
FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding.
[Code]
Yiweng Xie, Bo He, Junke Wang†, Xiangyu Zheng, Ziyi Ye, Zuxuan Wu.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2026.
|
OmniGen-AR: AutoRegressive Any-to-Image Generation.
[Code]
Junke Wang, Xun Wang, Qiushan Guo, Peize Sun, Weilin Huang, Zuxuan Wu, Yu-Gang Jiang.
Advances in Neural Information Processing Systems (NeurIPS), 2025.
|
OmniTracker: Unifying Object Tracking by Tracking-with-Detection.
Junke Wang*, Zuxuan Wu*, Dongdong Chen, Chong Luo, Xiyang Dai, Lu Yuan, Yu-Gang Jiang.
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025.
|
Fighting Malicious Media Data: A Survey on Tampering Detection and Deepfake Detection.
Junke Wang, Zhenxin Li, Chao Zhang, Jingjing Chen, Zuxuan Wu, Larry S. Davis, Yu-Gang
Jiang.
Proceedings of the IEEE, 2025.
|
OmniTokenizer: A Joint Image-Video Tokenizer for Visual Generation.
[Code]
Junke Wang, Yi Jiang, Zehuan Yuan, Binyue Peng, Zuxuan Wu, Yu-Gang Jiang.
Advances in Neural Information Processing Systems (NeurIPS), 2024.
|
OmniVid: A Generative Framework for Universal Video Understanding.
[Code]
Junke Wang, Dongdong Chen, Chong Luo, Bo He, Lu Yuan, Zuxuan Wu, Yu-Gang Jiang.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
|
Look Before You Match: Instance Understanding Matters in Video Object Segmentation.
Junke Wang, Dongdong Chen, Zuxuan Wu, Chong Luo, Chuanxin Tang, Xiyang Dai, Yucheng Zhao,
Yujia Xie, Lu Yuan, Yu-Gang Jiang.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
|
OmniVL: One Foundation Model for Image-Language and Video-Language Tasks.
Junke Wang, Dongdong Chen, Zuxuan Wu, Chong Luo, Luowei Zhou, Yucheng Zhao,
Yujia Xie, Ce
Liu, Yu-Gang Jiang, Lu Yuan.
Advances in Neural Information Processing Systems (NeurIPS), 2022.
|
Efficient Video Transformers with Spatial-Temporal Token Selection.
[Code]
Junke Wang*, Xitong Yang*, Hengduo Li, Zuxuan Wu, Yu-Gang Jiang.
European Conference on Computer Vision (ECCV), 2022.
|
ObjectFormer for Image Manipulation Detection and Localization.
[Code]
Junke Wang, Zuxuan Wu, Jingjing Chen, Xintong Han, Abhinav Shrivastava, Yu-Gang Jiang,
Ser-Nam Li.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
|
M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection.
[Code]
Junke Wang, Zuxuan Wu, Wenhao Ouyang, Xintong Han, Jingjing Chen, Ser-Nam Lim, Yu-Gang Jiang.
International Conference on Multimedia Retrieval (ICMR), 2022.
|
FT-TDR: Frequency-Guided Transformer and Top-Down Refinement Network for Blind Face Inpainting.
Junke Wang, Shaoxiang Chen, Zuxuan Wu, Yu-Gang Jiang.
IEEE Transactions on Multimedia (TMM), 2022.
|
|
Services
Conference Reviewer for CVPR, ICCV, ICML, NeurIPS, ICLR, ECCV, etal.
Journal Reviewer for TPAMI, TIP, IJCV, etal.
|
|