I’m Ruoyu Chen (陈若愚), a Ph.D. candidate in Institute of Information Engineering, Chinese Academy of Sciences (IIE, CAS), under the tutelage of Prof. Xiaochun Cao (currently the Dean of School of Cyber Science and Technology, Sun Yat-sen University) and Hua Zhang (Professor at IIE, CAS). I received my B.E. degree from Northeastern University, China. My research interests include Interpretable AI and Foundation Model.
I’m committed to making XAI meaningful and actually helping us with AI systems. Including in the model testing phase, building an interpretation method for debugging the model to help us discover potential biases and errors in the model. In the model training phase, specific defects are discovered through interpretability, and a reasonable feedback mechanism is designed to enable the model to automatically repair errors to improve model performance and generalization, while making the training process transparent. In the model deployment phase, improve the explanation of the model and study the human-computer interaction or AI agent processes in dynamic environments.
I am actively seeking a postdoctoral position and would welcome the opportunity to discuss potential collaborations, especially in the fields of Trustworthy AI and Foundation Model. Please don’t hesitate to get in touch if my research interests you.
🔥 News
- 2025.04.09: 🎉🎉 One paper is accepted by T-PAMI!
- 2025.04.06: 🎉🎉 VPS was selected as the Highlight paper of CVPR 2025.
- 2025.02.28: One paper is accepted by CVPR 2025!
- 2024.01.16: One paper is accepted by ICLR 2024 Oral!
- 2023.05: Created a new home page.
📝 Selected Publications
Generalized Semantic Contrastive Learning via Embedding Side Information for Few-Shot Object Detection
Ruoyu Chen, Hua Zhang, Jingzhi Li, Li Liu, Zhen Huang, and Xiaochun Cao
/ IEEE Transactions on Pattern Analysis and Machine Intelligence,
Impact factor: 20.8
Less is More: Efficient Black-box Attribution via Minimal Interpretable Subset Selection
Ruoyu Chen, Siyuan Liang, Jingzhi Li, Shiming Liu, Li Liu, Hua Zhang, and Xiaochun Cao
/ Preprint,
Black-box attribution.
Interpreting Object-level Foundation Models via Visual Precision Search
Ruoyu Chen, Siyuan Liang, Jingzhi Li, Shiming Liu, Maosen Li, Zheng Huang, Hua Zhang, and Xiaochun Cao
/ IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR 2025),
Highlight paper (387/13008, 2.98%)
Less is More: Fewer Interpretable Region via Submodular Subset Selection
Ruoyu Chen, Hua Zhang, Siyuan Liang, Jingzhi Li, and Xiaochun Cao
Code 
/ Paper / Slide / Poster / AI Time Presentation
International Conference on Learning Representations (ICLR),
Oral (85/7262, 1.16%)
Sim2Word: Explaining Similarity with Representative Attribute Words via Counterfactual Explanations
Ruoyu Chen, Jingzhi Li, Hua Zhang, Changchong Sheng, Li Liu, and Xiaochun Cao
/ ACM Transactions on Multimedia Computing, Communications, and Applications,
Impact factor: 5.1
Online Estimating Weight of White Pekin Duck Carcass by Computer Vision
Ruoyu Chen, Yuliang Zhao, Yongliang Yang, Shuyu Wang, Lianjiang Li, Xiaopeng Sha, Lianqing Liu, Guanglie Zhang and Wen Jung Li
/ Poultry Science (Top Journal in Agricultural and Biological Sciences),
Impact factor: 4.4📝 Collaboration Papers
Beyond Progress Measures: Theoretical Insights into the Mechanism of Grokking
Zihan Gu$^\dagger$, Ruoyu Chen$^{\dagger}$, Hua Zhang, Yue Hu, and Xiaochun Cao ($\dagger$: Equal contribution)
/ Preprint,
Grokking Mechanism.
Unpacking Positional Encoding in Transformers: A Spectral Analysis of Content-Position Coupling
Zihan Gu, Han Zhang, Ruoyu Chen, Yue Hu, Hua Zhang
Preprint,
Positional Encoding Mechanism.
FaceInsight: A Multimodal Large Language Model for Face Perception
Jingzhi Li, Changjiang Luo, Ruoyu Chen, Hua Zhang, Wenqi Ren, Jianhou Gan, and Xiaochun Cao
Preprint,
MLLMs for Face Perception.
Object Detectors in the Open Environment: Challenges, Solutions, and Outlook
Siyuan Liang, Wei Wang$^\dagger$, Ruoyu Chen$^{\dagger}$, Aishan Liu, Boxi Wu, Ee-Chien Chang, Xiaochun Cao, and Dacheng Tao ($\dagger$: Equal contribution)
Preprint 2024,
Survey paper
Identity-Preserving Face Anonymization via Adaptively Facial Attributes Obfuscation
Jingzhi Li, Lutong Han, Ruoyu Chen, Hua Zhang, Bing Han, Lili Wang, and Xiaochun Cao
/ 
ACM MM 2021, Oral
- More papers are being submitted, or please visit my Google Scholar to view all papers.
🎖 Honors and Awards
- 2020.12 China National Scholarship, Ministry of Education of the People’s Republic of China (Top 1.5%), NEU.
- 2019.12 China National Scholarship, Ministry of Education of the People’s Republic of China (Top 1.5%), NEU.
- 2018.12 China National Scholarship, Ministry of Education of the People’s Republic of China (Top 1.5%), NEU.
📖 Educations
- 2021.09 - 2026.06 (excepted),
School of Cyberspace Security, University of Chinese Academy of Sciences, China.
Ph.D. Candidate - 2017.09 - 2021.06,
School of Control Engineering, Northeastern University, China.
Bachelor of Engineering
Ranking: 2/119
🎤 Talking and Teaching
- 2024.6.27 Share a talk with Tokyo Institute of Technology online: Interpretation of the Foundation Model: Concepts, Challenges, and Applications
- 2024.5.10 Give an oral presentation in Vienna at ICLR 24 conference (Slide)
- 2024.2.28 Share a ICLR 24 paper “Less is More: Fewer Interpretable Region via Submodular Subset Selection” at AI Time
- 2023.12.26 Taught the undergraduate course “Explainable Artificial Intelligence” at Shenzhen Campus of Sun Yat-sen University
- 2023.10.20 Share a technical review “Survey on the interpretability of foundation models” at Huawei
💬 Professional Service
Journal Reviewer
- IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI) x11
- IEEE Transactions on Biometrics, Behavior, and Identity Science (T-BIOM) x3
Conference Reviewer
- CVPR 23, 24
- ICLR 24
- NeurIPS 23, 24
- ICML 23, 24
- ICCV 23
- ECCV 22, 24