Hi, I am a PhD student at the National University of Singapore, supervised by Professor Kenji Kawaguchi, driven by a passion for advancing the frontiers of artificial intelligence. My research explores the intersection of efficiency, controllability, and explainability in AI systems.
I started my research journey by delving into explainability—unpacking the “black box” of neural networks to reveal how they make decisions. Over time, my interests have grown broader. Today, I work on making AI not only more transparent but also smarter in its resource use and more controllable in its outputs, empowering us to shape these systems with precision and trust.
Prior to my PhD study, I finished my master’s at the Technical University of Munich, Germany. I spent a great time doing research at CAMP (Chair for Computer Aided Medical Procedures & Augmented Reality) under the supervision of Professor Nassir Navab. At CAMP, my research topic was about explainable AI and how to apply explanation methods on neural networks trained for medical purposes.
I did my bachelor’s study at RWTH Aachen University in Germany. My final year thesis and research at the senior stage is about hardware security (logic encryption on microprocessors), supervised by Professor Rainer Leupers. I spent one year with the group working on my thesis and later as an RA.
My vision is to contribute to the advancement of AI as a transformative tool that enhances the quality of life and addresses meaningful challenges for humanity. I aspire to develop AI systems that are safe, reliable, and impactful, fostering solutions that promote progress and well-being on a global scale. During my part-time, I enjoy traveling to different places, especially relaxing beside the sea, and exploring the other side of our planet through diving.
🔥 News
- 2024.08: 🎉🎉 I finished my internship at American Express Innovation Lab in Singapore.
📄 Preprint
Effortless Efficiency: Low-Cost Pruning of Diffusion Models
Yang Zhang, Er Jin, Yanfei Dong, Ashkan Khakzar, Philip Torr, Johannes Stegmaier, Kenji Kawaguchi
Diffusion model pruning by neural mask optimization.
Keywords: diffusion model, structural pruning
Project homepage: Here. Demo: Here at Huggingface Space
AttributionLab: Faithfulness of Feature Attribution Under Controllable Environments
Yang Zhang, Yawei Li, Mina Rezaei, Bernd Bischl, Philip Torr, Ashkan Khakzar, Kenji Kawaguchi
A feature attribution evaluation framework that leverages synthetic models and data to ensure correct ground truth knowledge.
Keywords: feature attribution, evaluation metric, synthetic data
📝 Selected Publications
A Dual-Perspective Approach to Evaluating Feature Attribution Methods
Yawei Li, Yang Zhang, Kenji Kawaguchi, Ashkan Khakzar, Bernd Bischl, Mina Rezaei
An feature attribution evaluation framework that evaluates the soundness and completeness of feature attribution methods.
Keywords: feature attribution, evaluation metric
Enhancing Semantic Fidelity in Text-to-Image Synthesis: Attention Regulation in Diffusion Models
Yang Zhang, Teoh Tze Tzun, Lim Wei Hern, Tiviatis Sim, Kenji Kawaguchi
An optimization-based approach to enhance the semantic fidelity of text-to-image diffusion models by regulating the attention maps in diffusion models.
Keywords: diffusion model, attention edit
Fine-grained neural network explanation by identifying input features with predictive information
Yang Zhang, Ashkan Khakzar, Yawei Li, Azade Farshad, Seong Tae Kim, Nassir Navab
Fine-grained neural network explanation by attributing input features with predictive information.
Keywords: feature attribution, information bottleneck
- Memory-Efficient Gradient Unrolling for Large-Scale Bi-level Optimization, Qianli Shen, Yezhen Wang, Zhouhao Yang, Xiang Li, Haonan Wang, Yang Zhang, Jonathan Scarlett, Zhanxing Zhu, Kenji Kawaguchi, NeurIPS 2024
- The stronger the diffusion model, the easier the backdoor: Data poisoning to induce copyright breaches without adjusting finetuning pipeline, Haonan Wang, Qianli Shen, Yao Tong, Yang Zhang, Kenji Kawaguchi, ICML 2024
- Explaining covid-19 and thoracic pathology model predictions by identifying informative input features, Ashkan Khakzar, Yang Zhang, Wejdene Mansour, Yuezhi Cai, Yawei Li, Yucheng Zhang, Seong Tae Kim, Nassir Navab, MICCAI 2021
💻 Internships
- 2024.05 - 2024.08, Data Scientist Intern, American Express Decision Science, Singapore, Singapore.
- 2022.03 - 2022.09, Research Intern, Microsoft Research Asia, Beijing, China.
📖 Educations
- Ph.D. student in Computer Science, School of Computing, National University of Singapore, Singapore, Singapore.
- M.Sc. in Robotics, Cognition, Intelligence, Department of Computer Science, Technical University of Munich, Munich, Germany.
- B.Sc. in Computer Engineering, Department of Electrical Engineering. RWTH Aachen University, Aachen, Germany.
💬 Academic Service
- Reviewer for CVPR 2024, 2025, ACM MM 2024, Neurips 2024, AAAI 2025, ICLR 2025