Ce Zhang 张策

About Me

Hi there! I’m Ce Zhang. I am currently a first-year PhD student in the Robotics Institute at Carnegie Mellon University (CMU).

Previously, I obtained my M.Sc. in Machine Learning in the Machine Learning Department at CMU, advised by Prof. Katia Sycara. Prior to this, I received my B.Eng. in Communication Engineering from Southern University of Science and Technology (SUSTech), where I worked under the supervision of Prof. Zhihai He. My current research interest lie in vision-language models and scene understanding.

Research Interests

I aim to enhance the adaptability and reliability of multi-modal models, pushing the boundaries of what they can achieve in real-world applications. My vision focuses on tackling pivotal challenges, including:

• How can we responsibly deploy multi-modal models while mitigating real-world societal concerns, such as mitigating misinformation and hallucinations, promoting fairness, and overcoming challenges with compositional reasoning?
• How can we enable a broader adaptability of these models to diverse real-world domains, such as robotics and healthcare, by leveraging resource-efficient fine-tuning, test-time adaptation, and continual learning?
• How can these models seamlessly interact with humans and their environments, such as advancing capabilities in human-object interaction detection, 3D visual reasoning, and video understanding?

News

• [Mar. 2025] I will be joining the Ph.D. in Robotics program at Carnegie Mellon University (CMU) in Fall 2025.
• [Jan. 2025] Our paper “Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models” is accepted to ICLR 2025.
• [Oct. 2024] Our paper “Enhancing Vision-Language Few-Shot Adaptation with Negative Learning” is accepted to WACV 2025.
• [Sep. 2024] Our paper “Dual Prototype Evolving for Test-Time Generalization of Vision-Language Models” is accepted to NeurIPS 2024.
• [Feb. 2024] Our paper “HiKER-SGG: Hierarchical Knowledge Enhanced Robust Scene Graph Generation” is accepted to CVPR 2024.
• [Sep. 2023] Our paper “Learning to Adapt CLIP for Few-Shot Monocular Depth Estimation” is accepted tco WACV 2024.

Publications

Honors and Awards

• Top 10 Summa Cum Laude Graduates, Southern University of Science and Technology, June 2023
• Chinese National Scholarship, Ministry of Education of the People’s Republic of China, Novermber 2022

Services

• Journal Reviewer: IJCV, IEEE TCSVT (>10), IEEE TIP, IEEE TMM, IEEE TAI, Neurocomputing, Knowledge-Based Systems, Pattern Recognition
• Conference Reviewer: NeurIPS ‘25, ICLR ‘25, ICCV ‘25, AAAI ‘25, WACV ‘25 (Outstanding Reviewer), ICASSP ‘25, BMVC ‘24 ‘25 (Outstanding Reviewer), ICME ‘24 ‘25, IJCNN ‘25, AVSS ‘25

 visitors since Sep 2023.

Powered by Jekyll and Minimal Light theme.