Researcher in Vision-Language Models & Agentic LLMs
I am a student researcher at X-D Lab advised by Yiyi Liao. I am also lucky to have collaboration with Matteo Poggi. Previously I obtained my B.Eng. degree in Automation from Zhejiang University in 2024.
CVPR 2025
Prometheus introduces a novel method for feed-forward scene-level 3D generation in seconds, harnessing pre-trained 2D priors for generalizable and efficient 3D synthesis.
CVPR 2025
ChronoDepth addresses the challenge of temporally consistent video depth estimation using video diffusion model priors.
CVPR 2024
HUGS utilizes 3D Gaussian Splatting for holistic urban scene understanding with only posed RGB images, achieving real-time rendering at 100 fps while modeling exposure, optical flow, semantics, and dynamic objects.
A personal configuration template for OpenClaw, an AI-powered automation system for eliminating repetitive digital work.
An AI-agent-compatible skill that transforms Notion into a personal knowledge management system using the PARA method.
My personal Claude Code configuration, skills, and workflow setup for AI-assisted software development.
