Xinkai Wang (王薪恺) — SII & SEU | Embodied AI, VLA, Video Models

About

I am currently a PhD student at the Shanghai Innovation Institute (SII), jointly trained with Southeast University (SEU). I am primarily advised by Prof. Lixin Yang and Prof. Cewu Lu, with Prof. Lifeng Zhu as my joint-training advisor. I am also working closely with Yutao Sun (THU) on unified multimodal architectures and autoregressive video generation. I got my B.Eng at the School of Instrument Science and Engineering, Southeast University in 2023. My research interests include Embodied AI & Robot Learning, Multimodal Understanding & Generation. I'm always looking for related collaboration. If you are interested in chatting with me, feel free to drop me an email.

Interests

Embodied AI & Robot Learning

OneWM ForceWAM Track4Action LaMP

Multimodal Understanding & Generation

AD-Loop

Human-AI Interaction & Haptics

Deep Cognition DipMe

Medical XR & Robot-Assisted Intervention

MRUCT VISTA

Education

Shanghai Innovation Institute

Ph.D. Candidate in Instrument Science and Engineering, Sep. 2025 - Now

Primary advisors: Prof. Lixin Yang and Prof. Cewu Lu · Joint training with Southeast University under Prof. Lifeng Zhu

Southeast University

M.Sc in Instrument Science and Engineering, Sep. 2023 - Feb. 2025

Supervisor: Prof. Lifeng Zhu

Southeast University

B.Eng in Measurement and Control Technology and Instrument, Sep. 2019 - Jun. 2023

Experiences

AI Intern @ Noematrix

Responsible for Ego Data Pre-Training.

Aug, 2026 - Present

Visiting Student @ ORoboSoul, SII

Focus on Embodied AI and VLA models.

Supervisor: Prof. Lixin Yang and Prof. Cewu Lu

Mar, 2025 - Aug, 2025

News

Recent updates

Oct, 2026 Two co-first-author papers, OneWM and ForceWAM, are now available on arXiv.
Aug, 2026 Our paper Track4Action is now available on arXiv.
Jun, 2026 One paper accepted to ECCV 2026 🇸🇪.
Feb, 2026 One paper accepted to ICLR 2026 🇧🇷.
Oct, 2025 As Team Leader, won 3rd Place in the IROS 2025 🇨🇳 InternUtopia & Real-World Challenge, Track 1 (Manipulation).
May, 2025 One paper accepted to ISMAR 2025 🇰🇷, and nominated for Best Poster Award.
Dec, 2024 One paper accepted to IEEE VR 2025 🇫🇷, and selected for Oral presentation.
Dec, 2023 Won Outstanding Award (Top 5%) in the 18th Challenge Cup National College Student Extracurricular Academic and Technological Works Competition.

Publications

* Equal contribution † Corresponding author

Can World and Action Modeling Benefit Each Other under Unified Causal Training?

Yue Su^*, Bokai Lin^*, Xinkai Wang^*, Chubin Zhang, Xiangnan Wu, Kuan Zhang, Weiyang Jin, Jiwen Yu, Yiming Li, Cewu Lu, Lixin Yang^†, Xihui Liu^†

HKU, SJTU, THU, SII, UCAS, IAAI Team

arXiv 2026

Website Paper

ForceWAM: Decoupling Future and Response Learning from Demo and Play Data for Contact-Rich Manipulation

Zijun Xu^*, Wendi Chen^*, Xinkai Wang^*, Jiachen Zou, Hongjin Chen, Hairuo Liu, Bokai Lin, Shihao Ma, Ke Ma, Zhiwei Zhang, Zhuohao Yan, Ke Wu, Haoming Song, Xueqi Li, Chuan Wen, Jieru Zhao, Wenchao Ding, Cewu Lu, Lixin Yang^†

FDU, SJTU, SEU, SII, SUSTech, Noematrix

arXiv 2026

Website Paper

Track4Action: Distilling World-Centric 3D Tracker into Vision-Language-Action Policies

Chenyi Wang^*, Xinkai Wang^*, Bokai Lin, Jialin Tian, Fucheng Zhang, Cewu Lu, Lixin Yang^†

ZJU, SJTU, SII, Noematrix

arXiv 2026

Website Paper

LaMP: Learning Vision-Language-Action Policy with 3D Scene Flow as Latent Motion Prior

Xinkai Wang, Chenyi Wang, Yifu Xu, Mingzhe Ye, Fucheng Zhang, Jialin Tian, Xinyu Zhan, Lifeng Zhu, Cewu Lu, Lixin Yang^†

SEU, SJTU, SII, Noematrix

ECCV 2026

Website Paper Code

Synergizing Understanding and Generation with Interleaved Analyzing-Drafting Thinking

Shengqiong Wu, Bobo Li, Xinkai Wang, Xiangtai Li, Lei Cui, Furu Wei, Shuicheng Yan, Hao Fei^†, Tat-Seng Chua

NUS, NTU, Microsoft Research

ICLR 2026

Website Paper

Interaction as Intelligence: Deep Research With Human-AI Partnership

Lyumanshan Ye, Xiaojie Cai, Xinkai Wang, Junfei Wang, Xiangkun Hu, Jiadi Su, Yang Nan, Sihan Wang, Bohan Zhang, Xiaoze Fan, Jinbin Luo, Yuxiang Zheng, Tianze Xu, Dayuan Fu, Yunze Wu, Pengrui Lu, Zengzhi Wang, Yiwei Qin, Zhen Huang, Yan Ma, Zhulin Hu, Haoyang Zou, Tiantian Mi, Yixin Ye, Ethan Chern, Pengfei Liu^†

SII-GAIR, SJTU

arXiv 2025

Website Paper