HE Tianlun (Allen)

Embodied AI · Robot Learning · Computer Vision

Portrait of HE Tianlun (Allen)

HE Tianlun (Allen)

The Hong Kong University of Science and Technology香港科技大学

📍 Hong Kong, China中国 · 香港

📧 theaj@connect.ust.hk

🐙 github.com/Allenhetl

💬 WeChat微信: 18216000982

👋 About Me

Hi! I’m HE Tianlun (Allen), a Year-3 undergraduate at The Hong Kong University of Science and Technology (HKUST) (expected graduation June 2027), majoring in Integrative Systems and Design with a Minor in Robotics.

My research interests center on Embodied AI — particularly vision-language-action (VLA) models, memory-augmented policies, world models for robotic manipulation, and sim-to-real reinforcement learning. I am currently working on Memory-Augmented VLA, and interning as a VLA Algorithm Intern at Mondo Tech (妙动科技), where I train and deploy VLA policies for humanoid whole-body control.

I’m actively looking for research internships and PhD opportunities in embodied AI and robot learning. If my experience interests you, I’d be happy to connect.

Outside research, I build LEGO Technic sets, take photos, and travel when I can.

👋 关于我

你好!我是 贺天伦(Allen)香港科技大学大三学生(预计 2027 年 6 月毕业),主修综合系统与设计,辅修机器人方向

我的研究方向集中在具身智能(Embodied AI),包括视觉-语言-动作(VLA)模型、记忆增强策略、机器人操作的世界模型,以及 sim-to-real 强化学习。目前在做 Memory-Augmented VLA;同时在 妙动科技(Mondo Tech)担任 VLA 算法实习生,负责面向人形机器人全身控制的 VLA 策略训练与真机部署。

我正在寻找研究实习博士项目机会。如果我的经历能引起你的兴趣,欢迎随时联系。

科研之余,我喜欢拼乐高 Technic、拍照和旅行。

news

May 15, 2026 💼 Joined Mondo Tech (妙动科技) in Shenzhen as a VLA Algorithm Intern, working on vision-language-action models for robotic manipulation.💼 加入深圳妙动科技 (Mondo Tech),担任 VLA 算法实习生,从事面向机器人操作的视觉-语言-动作模型研发。
Apr 15, 2026 🎉 Our paper MVISTA-4D: View-Consistent 4D World Model with Test-Time Action Inference for Robotic Manipulation has been accepted to ICML 2026! arXiv🎉 论文 MVISTA-4D: View-Consistent 4D World Model with Test-Time Action Inference for Robotic ManipulationICML 2026 接收arXiv
Feb 15, 2026 🚀 Started Memory-Augmented VLA (URP project) — designing a subtask-aware framework with VAE-based compression for long-horizon manipulation.🚀 启动记忆增强 VLA(本科生研究计划):基于 VAE 的子任务压缩框架,缓解长时序任务的上下文爆炸。
Sep 01, 2025 🤖 Joined the Undergraduate Research Program on hydraulic soft robotic arms under the supervision of Prof. ROB.🤖 加入 ROB 教授指导的本科生研究计划,参与液压软体机械臂课题。
Apr 15, 2025 🏆 Won the Champion title of RoboMaster 2025 University League with HKUST Team ENTERPRIZE.🏆 与 HKUST ENTERPRIZE 战队夺得 RoboMaster 2025 高校联盟赛冠军
Sep 01, 2024 🎓 Awarded the HKSAR Government Scholarship Fund — Talent Development Scholarship (2024/25).🎓 获香港特别行政区政府奖学基金 — Talent Development Scholarship(2024/25 学年)。
Aug 15, 2024 🥇 Won the Champion title of RoboMaster 2024 RMUC International Region with HKUST Team ENTERPRIZE.🥇 与 HKUST ENTERPRIZE 战队夺得 RoboMaster 2024 RMUC 国际赛区冠军
Nov 01, 2023 🛠️ Joined HKUST RoboMaster Team ENTERPRIZE as a mechanical member (2023 – 2025).🛠️ 加入 HKUST RoboMaster Enterprize 战队,担任机械成员(2023 – 2025)。
Jun 15, 2023 🎓 Joined HKUST, Integrative Systems and Design (ISD) with a Minor in Robotics.🎓 入读香港科技大学综合系统与设计(ISD)专业,辅修机器人方向。

selected publications

  1. ICML
    MVISTA-4D: View-Consistent 4D World Model with Test-Time Action Inference for Robotic Manipulation
    Jiaxu Wang, Yicheng Jiang, Tianlun He, and 8 more authors
    In Proceedings of the 43rd International Conference on Machine Learning (ICML), Apr 2026