Kangning Cui

Kangning Cui

Assistant Professor of Data ScienceCity University of Hong Kong (Dongguan)

Office: AC4-422

I develop efficient and reliable machine learning methods for language models, geospatial data, and medical images and signals. My work focuses on model compression, interpretability, and learning with limited labels.

PhD and research assistant opportunities: join the group.

Email CityU-DG Profile GitHub LinkedIn Google Scholar ORCID Scopus
Kangning Cui

Recent News

[09/2026]
LowRankArena accepted at NeurIPS'26, ED Track, Poster.
[09/2026]
Extending ECG foundation models to longer recordings accepted at NeurIPS'26, Poster.
[08/2026]
DPQ: preserving uncertainty in quantized language models accepted at EMNLP'26, Findings.
[08/2026]
GeoTALOS, our geoscience platform, accepted at SIGSPATIAL'26.
Earlier news
[06/2026]
The paper "JassidNet as a Quantization-Aware Lightweight Phenotyping Framework for High-Throughput Cotton Jassid (Amrasca biguttula) Detection and Counting Toward Objective Resistance Screening" was accepted by Ind. Crops Prod..
[04/2026]
The paper "DecodeShare: Tracing the Shared Subspace of LLM Decode-Time Decisions" was accepted by ICML 2026 (Spotlight).
[04/2026]
The paper "AwakeForest: An Interactive Geospatial Platform for Large-Scale Forest Imagery" was accepted by IJCAI Demo 2026.
[04/2026]
The paper "Learning Where to Embed: Noise-Aware Positional Embedding for Query Retrieval in Small-Object Detection" was accepted by ICMR 2026.
All research
LowRankArena overview: evaluation challenges, a standardized SVD-compression platform, and findings on accuracy and inference efficiency

Language Models

LowRankArena: evaluating LLM compression

LowRankArena compares SVD-based language-model compression using shared tasks, compression budgets, and inference measurements.

NeurIPS'26, ED Track, Poster

PRISM workflow: aerial forest imagery, palm detection, segmentation, and geographic mapping

Geospatial AI

Mapping palms in tropical forests

PRISM turns large aerial images into geographic maps of individual palms, combining detection, segmentation, and confidence calibration.

IJCAI'25, Oral

S2DL workflow: hyperspectral data, superpixels, diffusion geometry, and clustering without labels

Hyperspectral Learning

S2DL: clustering hyperspectral images

S2DL combines superpixels and diffusion geometry to cluster hyperspectral images without labels while reducing computational cost.

TGRS'24

Join the Group

PhD opportunities at CityU and research assistant positions at CityU-DG. For CityU-DG projects, feel free to email me to discuss your interests.

Application details

Appointments

2026–Present
City University of Hong Kong (Dongguan)
Assistant Professor of Data Science
2025–Present
Andrew Sabin Family Center for Environment and Sustainability
Robert J. Plemmons Fellow
2025–2026
Wake Forest University
Research Assistant Professor, Department of Computer Science

Education

2021–2025
Ph.D. in Mathematics
City University of Hong Kong & Hong Kong Centre for Cerebro-Cardiovascular Health Engineering

Research in image and signal processing for medicine and remote sensing, advised by Prof. Raymond H. Chan and Prof. Jean-Michel Morel.

2024–2025
Visiting DPhil Student in Biomedical Engineering
University of Oxford

Research in ultrasound video analysis and biomedical imaging, advised by Prof. Alison Noble.

2017–2021
B.S. in Computing Mathematics
City University of Hong Kong

Training in computing mathematics with First Class Honours and a minor in Computer Science.

2020
Exchange Student, Applied Mathematics
Delft University of Technology

Exchange study in applied mathematics in the Netherlands.