Yue Chang | 常越

I am currently a first-year PhD student at The Hong Kong University of Science and Technology (Guangzhou), advised by Prof. Sihong Xie. I received my B.Eng. degree in Automation from Harbin Institute of Technology, Shenzhen.

Email / CV / GitHub / Google Scholar / Blog / WeChat

Email copied

Education

The Hong Kong University of Science and Technology (Guangzhou) 2025.03 - now

PhD Student | Advisor: Prof. Sihong Xie

Harbin Institute of Technology, Shenzhen 2021.09 - 2025.07

B.Eng. in Automation

Research

My research focuses on spatiotemporal perception and memory for embodied intelligence, with a current emphasis on 3D scene graphs (3DSGs). I also work on vision-language navigation (VLN), 3D reconstruction, and multimodal LLM (MLLMs).

In the Year of 2026
	PORTER: Edge-Cloud Residency for Persistent 3D Scene Graph Memory Yue Chang, Yifan Tian, Jiajing Peng, Dazhi Huang, Rufeng Chen, Zhaofan Zhang, Li Chen, Sihong Xie arXiv preprint arXiv
	Just-In-Time Scene Graph Growth: Combating Perceptual Saturation in Long-Horizon Robotics Yue Chang, Rufeng Chen, Yifan Tian, Dazhi Huang, Zhaofan Zhang, Yi Chen, Wenze Zhang, Li Chen, Hui Xiong, Sihong Xie arXiv preprint arXiv
	LH-AVLN: A Benchmark for Long-Horizon Audio-Visual-Language Navigation Rufeng Chen, Yue Chang, Zili Shao, Zhaofan Zhang, Hechang Chen, Hui Xiong, Sihong Xie arXiv preprint arXiv
	RAG-3DSG: Enhancing 3D Scene Graphs with Re-Shot Guided Retrieval-Augmented Generation Yue Chang, Rufeng Chen, Zhaofan Zhang, Yi Chen, Yifan Tian, Sihong Xie Accepted by ECCV 2026 arXiv / Code / Video
	PSG-Nav: Probabilistic Scene Graph Navigation via Multiverse Decision Making Rufeng Chen, Yue Chang, Xiaqiang Tang, Hechang Chen, Sihong Xie Accepted by ICML 2026 arXiv / Code
In the Year of 2024
	A Unified Hallucination Mitigation Framework for Large Vision-Language Models Yue Chang, Liqiang Jing, Xiaopeng Zhang, Zhang Yue Accepted by TMLR* arXiv / Code
In the Year of 2023
	A LIO System with Front/Back-End Odometry and Loop-Detection Yue Chang Used by Competition Team (Private) A fusion odometry: It uses LIO as the front-end odometry, uses imu pre-integration to fill the frames, and uses ScanContext as the spatial descriptor for back-end loop detection.

Academic Service

Journal Reviewer: IEEE Transactions on Robotics (T-RO), IEEE Transactions on Image Processing (TIP).

Conference Reviewer: ICRA, ICLR, ECCV, ICML.

Award

Student Representative, HKUST(GZ) PhD Admission Seminar 2026
National First Prize, Infantry Catogory of Robomaster Super Competition 2023
National Second Prize, Robomaster Super Team Competition 2023
National Second Prize, Mineral Cup Logistics Technology Creativity Competition 2023
National Second Prize, China Robot and Artificial Intelligence Competition 2023
National Third Prize, BOTEC international intelligent Robot Technical Challenge 2022
National Third Prize, Sentinel Catogory of Robomaster Super Competition 2023