About me (Curriculum Vitae)

Dr. Xiang Li is a Lecturer (Assistant Professor) in the Department of Computer Science at the University of Reading. His research focuses on multimodal large language models, computer vision, and remote sensing. Dr. Li has published over 50 papers in top conferences and journals in computer vision and remote sensing, such as CVPR, ICCV, NeurIPS, TVCG, and TGRS, with citations on Google Scholar and an h-index of 27. Dr. Li contributed to the well-known MiniGPT-4 project, with over 3000 citations on Google Scholar. Dr. Li co-organized the 1st and 2nd workshops on Compositional 3D Vision and the 3DCoMPaT dataset challenge at IEEE CVPR 2023 and 2024. Additionally, he serves as Guest Editor for the special issue Vision-Language Models in Remote Sensing” at IEEE GRSM.

🔥Hiring PhD Students and Interns: Welcome enthusiastic students passionate about computer vision, multimodal large language models, and remote sensing to apply for PhD positions and internships.

🔥 News

[05/2025] We release REOBench: A benchmark for evaluating the robustness of Earth observation foundation models.
[05/2025] I gave a seminar talk “Large Vision Language Models in Remote Sensing: Datasets and Models” at the Data Assimilation Research Center.
[03/2025] RSGPT was accepted for publication at ISPRS JPRS. RSGPT is the first attempt at GPT-based MLLMs in remote sensing.
[03/2025] Stable-SPAM was accepted for publication at COLM.
[12/2024] Opening position: One fully funded PhD position in AI for Bioversity (closed).
[10/2024] I joined the University of Reading as a Lecturer in Computer Science.
[09/2024] Two papers accepted at NeruIPS 2024 Datasets and Benchmarks Track VRSBench and 3DCoMPaT200.
[07/2024] One paper on 3D LLM (Uni3DL) was accepted by ECCV 2024.
[07/2024] I served as the Guest Editor for the special issue “Vision-Language Models in Remote Sensing” at IEEE GRSM.
[06/2024] I’m co-organizing the C3DV 2024: 2nd Workshop on Compositional 3D Vision at CVPR 2024.
[05/2024] One paper on few-shot object detection (InfRS) was accepted by TGRS.
[05/2024] One paper on CO₂ mapping was published at JAG.
[04/2024] Our survey paper Vision-Language Models in Remote Sensing was published at IEEE GRSM.
[03/2024] I gave a talk “AI for Earth Observation” at Prof. Matthew McCabe’s lab.
[11/2023] RS-CLIP was published at JAG.
[10/2023] We released MiniGPT-v2.
[07/2023] We released RSGPT, the first GPT-based multimodal LLM in remote sensing.
[05/2023] We released the survey paper [Vision-Language Models in Remote Sensing(https://arxiv.org/abs/2305.05726).
[04/2023] We released MiniGPT-4.

Selected Publications

[# denotes equal contribution, * denotes corresponding author]

	MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models. Deyao Zhu, Jun Chen, Xiaoqian Shen, Xiang Li, Mohamed Elhoseiny. ICLR, 2024 [project] [paper] [code] [huggingface demo]
	MiniGPT-v2: Large Language Model as a Unified Interface for Vision-Language Multi-task Learning. Jun Chen, Deyao Zhu, Xiaoqian Shen, Xiang Li, Zechun Liu, Pengchuan Zhang, Raghuraman Krishnamoorthi, Vikas Chandra, Yunyang Xiong, Mohamed Elhoseiny. arxiv, 2023 [project] [paper] [code] [huggingface demo]
	Vision-Language Models in Remote Sensing: Current Progress and Future Trends. Xiang Li, Congcong Wen, Yuan Hu, Zhengpeng Yuan, Xiao Xiang Zhu. IEEE Geoscience and Remote Sensing Magazine (GRSM)*, 2024 [paper]
	RSGPT: A Remote Sensing Vision Language Model and Benchmark. Yuan Hu, Jianlong Yuan, Congcong Wen, Xiaonan Lu, Xiang Li. ISPRS JPRS*, 2025 (arXiv 2023) [project] [paper] [code]
	RS-CLIP: Zero Shot Remote Sensing Scene Classification via Contrastive Vision-Language Supervision. Xiang Li, Xiang Li, Congcong Wen, Nan Zhou. International Journal of Applied Earth Observation and Geoinformation (JAG), 2023 [project] [paper] [code]
	Few-shot Object Detection on Remote Sensing Images. Xiang Li^#, Jingyu Deng^#, Yi Fang. TGRS, 2021 [paper] [code]
	Few-shot Learning of Part-specific Probability Space for 3D Shape Segmentation. Lingjing Wang^#, Xiang Li^#, Yi Fang. CVPR, 2020 [paper] [code]
	Directionally Constrained Fully Convolutional Neural Network For Airborne Lidar Point Cloud Classification. Congcong Wen, Lina Yang, Ling Peng, Xiang Li* ISPRS JPRS, 2020 [paper] [code]

Honors and Awards

Outstanding Reviewer for ICCV 2021.
Postdoc Non-travel Award, NYUAD 2020 & 2021.
National Scholarship, University of Chinese Academy of Sciences, 2018
China Scholarship Council (CSC) scholarship, 2017
Director’s Fund of RADI, 2017
Seagate Scholarship, Wuhan University, 2012
National Scholarship, Wuhan University, 2011

Xiang Li

Selected Publications

MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models.

MiniGPT-v2: Large Language Model as a Unified Interface for Vision-Language Multi-task Learning.

Vision-Language Models in Remote Sensing: Current Progress and Future Trends.

RSGPT: A Remote Sensing Vision Language Model and Benchmark.

RS-CLIP: Zero Shot Remote Sensing Scene Classification via Contrastive Vision-Language Supervision.

Few-shot Object Detection on Remote Sensing Images.

Few-shot Learning of Part-specific Probability Space for 3D Shape Segmentation.

Directionally Constrained Fully Convolutional Neural Network For Airborne Lidar Point Cloud Classification.

Honors and Awards