About
I am an MS student at the National University of Singapore (NUS), working with Show Lab and advised by Prof. Mike Zheng Shou. Prior to NUS, I received my B.E. in Computer Science from Zhejiang University, where I was advised by Prof. Juncheng Li.
My research lies at the intersection of embodied AI, multimodal large language models, and 3D scene understanding. I am passionate about building AI systems that can perceive, reason about, and act in the physical world. I have co-authored papers on mitigating hallucinations in LVLMs and developing real-time scene understanding systems integrated into wearable smart-vision aids.
News
2026.09
Survey Released Awesome Multimodal Embodied Agents: From Computer-Use to Robot-Use — PAPAV (Perceive → Anticipate → Plan → Act → Verify), a capability-centric map of the Agent × Robotics landscape, with an analysis of 64 benchmarks.
2026.03
Joined Alibaba Tongyi Lab (Wanxiang) as Data & Algorithm Intern, working on universal inverse dynamics models and cross-domain action alignment.
2025.11
Joined ACE Robotics as Embodied Intelligence Intern, working on Qwen3-VL fine-tuning and spatial intelligence data pipelines.
2025.05
Paper Causal-LLaVA published on arXiv — mitigating hallucinations in multimodal LLMs via causal disentanglement.
2025.04
Joined Unitree Robotics as Assistant Engineer, developing guidance systems for the Unitree G1 humanoid robot.
Publications
Awesome Multimodal Embodied Agents: From Computer-Use to Robot-Use
Survey 2026
Causal-LLaVA: Causal Disentanglement for Mitigating Hallucination in Multimodal Large Language Models
arXiv 2025