Guangrui Li @ ReLER

Guangrui Li 李光睿

I am currently an applied scientist at Amazon Rufus, building LLM-powered Shopping Agent, in Palo Alto. My recent work focuses on building scalable and elestic multi-agent system in real production.

I obtained my Ph.D. degree from ReLER Lab, Australian Artificial Intelligence Institute, University of Technology Sydney, under the supervision of Prof. Yi Yang. During my Ph.D. study, I am fortunate to be sequentially mentored by Prof. Yunchao Wei, and Dr. Guoliang Kang.

Email / CV / Google Scholar / Github

News

[July 2025] Finished San Francisco Half Marathon in 02:03:03.

[July 2024] One paper accepted by ECCV2024.

[Feb. 2024] One paper accepted by CVPR2024.

[Sep. 2023] Finished Sydney Marathon in 04:22:10.

[Aug. 2023] Finished the Iconic City2Surf in 01:18:25.

[June 2023] Deliver PhD oral presentation (Candidate Assessment).

[Mar. 2023] Awarded Google Gift Fund.

[Mar. 2023] One paper accepted by CVPR2023.

[Jan. 2023] One paper accepted by ICLR2023.

[Sep. 2022] Finished Sydney Half Marathon in 01:53:49.

[Mar. 2021] Two papers accepted by CVPR2021.

[June 2020] One paper accepted by ECCV2020.

[Sep. 2019] Finished Sydney Marathon in 04:09:24.

Research Interests

[Agentic AI]: Devising effective evolving orchestration strategy for Multi-Agent System (MAS) in real production. Building up scalable and robust orchestration system that can accommodation dynamic evolution within MAS.

[Foundation Models]: Investigating effective fine-tuning strategy for LLM/MLLM, in terms of OOD generalization, and reasoning capabilties.

[3D Vision & Scene Understanding]: Perception and understanding with various visual structures, i.e., 2D/3D scene understanding (detection and segmentation), pose estimation, and sensor fusion (e.g., BEV in autonumous drinving).

Publications

	Robustness Preserving Fine-tuning using Neuron Importance Guangrui Li, Rahul Duggal, Aaditya Singh, Kaustav Kundu, Bing Shuai, Jonathan Wu ECCV, 2024 PDF
	Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation Guangrui Li CVPR, 2024 PDF
	Generalizable Visual Understanding with Deep Neural Networks Guangrui Li Ph.D. Thesis, 2023 PDF
	Adversarially Masking Synthetic to Mimic Real: Adaptive Noise Injection for Point Cloud Segmentation Adaptation Guangrui Li, Guoliang Kang, Xiaohan Wang, Yunchao Wei, Yi Yang CVPR, 2023 PDF
	Decompose to Generalize: Species-Generalized Animal Pose Estimation Guangrui Li, Yifan Sun, Zongxin Yang, Yi Yang ICLR, 2023 PDF / bibtex
	Domain Consensus Clustering for Universal Domain Adaptation Guangrui Li, Guoliang Kang, Yi Zhu, Yunchao Wei, Yi Yang CVPR, 2021 PDF / code / bibtex
	VSPW: A Large-scale Dataset for Video Scene Parsing in the Wild Jiaxu Miao, Yunchao Wei, Yu Wu, Chen Liang, Guangrui Li, Yi Yang CVPR, 2021 PDF/ Project Page / bibtex
	Content-Consistent Matching for Domain Adaptive Semantic Segmentation Guangrui Li, Guoliang Kang, Wu Liu, Yunchao Wei, Yi Yang (* Equal Contribution) ECCV, 2020 PDF / code / bibtex

Misc

Reviewers for: TPAMI, TKDE, TIP, KBS.

Running, Hiking, Swimming.

Website design from Jon Barron.