Ning Zhu 朱凝
Generative Modeling · Computer Vision · Medical AI
University of Electronic Science and Technology of China
About Me
Welcome to Ning Zhu’s Homepage! I am an incoming M.S. student in Electrical Engineering at Stanford University. I am currently a senior undergraduate at Glasgow College, University of Electronic Science and Technology of China (UESTC). I have been fortunate to be advised by Prof. Guotai Wang at UESTC and to collaborate with Prof. Han Liu and Jerry Yao-Chieh Hu at Northwestern University. My research interests lie in generative models, computer vision, and medical image analysis , particularly in building data-efficient and reliable models that bridge controlled benchmarks and real-world deployment. I am always happy to chat about research — feel free to drop me an email.
News
- May 2026 Our paper "Stabilizing Temporal Inference Dynamics for Online Surgical Phase Recognition" is early accepted by MICCAI 2026. 🎉
- Apr 2026 Started a research internship at Tencent AI focusing on Video Generation . 🐧
- Feb 2026 Admitted to Stanford University as an M.S. student in Electrical Engineering. 🌲
- Sep 2025 Our paper "High-Order Flow Matching: Unified Framework and Sharp Statistical Rates" is accepted by NeurIPS 2025.
- Sep 2025 Our paper "Deep Adaptive Wavelet Autoencoder for Unsupervised Motor Anomaly Detection" is accepted by Engineering Applications of Artificial Intelligence.
- Aug 2025 Released "MedCAL-Bench" on arXiv — a comprehensive benchmark on cold-start active learning with foundation models for medical image analysis.
- Jul 2025 Our paper "Adversarial Frequency Component Reconstruction Constraint for Helicopter Vibration Signal Anomaly Detection" is published in IEEE Transactions on Instrumentation and Measurement.
- May 2025 Our paper "CSAL-3D: Cold-Start Active Learning for 3D Medical Image Segmentation" is accepted by MICCAI 2025, and selected as Best Paper & Young Scientist Awards Shortlist. 🏆
- May 2025 Our paper "SUGFW: A SAM-Based Uncertainty-Guided Feature Weighting Framework for Cold Start Active Learning" is accepted by MICCAI 2025.
- May 2025 Our paper "Unsupervised Anomaly Detection for Aircraft PRSOV with Random Projection-Based Inner Product Prediction" is published in IEEE Transactions on Instrumentation and Measurement.
- Sep 2024 Our paper "An Adversarial Training Framework for Crystalline Silicon Solar Cells Anomaly Detection" is published in IEEE Transactions on Instrumentation and Measurement.
Education
- Sep 2026 — (expected) M.S. in Electrical Engineering Stanford University
- Sep 2022 — Jun 2026 B.Eng. in Electronic Information Engineering Glasgow College, University of Electronic Science and Technology of China (UESTC) Joint program with the University of Glasgow.
Research Interests
- Generative Models — flow matching, diffusion, image and video generation (both theoretical and practical approaches)
- AI4Medicine — medical image segmentation, surgical phase recognition
- Unsupervised Anomaly Detection — AI-powered signals processing, image anomaly detection
Selected Publications
-
-
MedCAL-Bench: A Comprehensive Benchmark on Cold-Start Active Learning with Foundation Models for Medical Image Analysis
arXiv preprint, arXiv:2508.03441, 2025
A unified benchmark systematically evaluating cold-start active learning strategies on top of foundation models for medical image analysis across diverse modalities.
-
CSAL-3D: Cold-Start Active Learning for 3D Medical Image Segmentation via SSL-Driven Uncertainty-Reinforced Diversity Sampling
Early accepted by MICCAI 2025 Early Accept 🏆 Best Paper & Young Scientist Awards Shortlist
An SSL-driven uncertainty-reinforced diversity sampling strategy that tackles the cold-start regime of 3D medical image segmentation under extremely limited annotation budgets.
-
Deep Adaptive Wavelet Autoencoder with Mutually Independent Empirical Cumulative Distribution for Unsupervised Motor Anomaly Detection
Engineering Applications of Artificial Intelligence, 2025
A deep adaptive wavelet autoencoder that leverages mutually independent empirical CDF modeling to detect motor anomalies without supervision.
-
Adversarial Frequency Component Reconstruction Constraint for Helicopter Vibration Signal Anomaly Detection: An Unsupervised Dual-Domain Approach
IEEE Transactions on Instrumentation and Measurement, 2025
An unsupervised dual-domain adversarial framework that reconstructs frequency components for vibration anomaly detection in helicopter signals.
-
Unsupervised Anomaly Detection for Aircraft PRSOV with Random Projection-Based Inner Product Prediction
IEEE Transactions on Instrumentation and Measurement, 2025
A random projection-based inner product prediction approach for unsupervised PRSOV anomaly detection in aircraft pneumatic systems.
-
-
An Adversarial Training Framework Based on Unsupervised Feature Reconstruction Constraints for Crystalline Silicon Solar Cells Anomaly Detection
IEEE Transactions on Instrumentation and Measurement, 2024
An adversarial training framework based on unsupervised feature reconstruction constraints for crystalline silicon solar cell anomaly detection.
Academic Services
- Reviewer: MICCAI 2026, Pattern Recognition, IEEE Transactions on Sustainable Energy
Teaching
- Machine Learning and Artificial Intelligence (Semester 2024-2025) — Teaching Assistant