avatar

Dingfan Chen

Research Group Leader
Max Planck Institute for Intelligent Systems (MPI-IS)
dingfan.chen(at)outlook.com


Dingfan Chen

I am currently a Research Group Leader in the Social Foundations of Computation Department, led by Prof. Dr. Moritz Hardt, at the Max Planck Institute for Intelligent Systems in Tübingen, Germany. I completed my Ph.D. in Computer Science at CISPA - Helmholtz Center for Information Security and Saarland University, under the supervision from Prof. Dr. Mario Fritz. Previously, I obtained my Master's degree in CS from Saarland University and Bachelor's degree in CS from University of Tübingen.

I am working on trustworthy and reliable machine learning – with a special focus on building privacy-preserving and data-efficient intelligent systems. My research interests include but are not limited to privacy, generative modeling, federated learning, and adversarial ML.

Publications [ Google Scholar ] [ Semantic Scholar ]

  1. ICML
    Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing
    Yuan Xin, Dingfan Chen, Michael Backes, Xiao Zhang
    International Conference on Machine Learning (ICML), 2025.
    PDF Code BibTex

  2. NeurIPS
    PrivAuditor: Benchmarking Data Protection Vulnerabilities in LLM Adaptation Techniques
    Derui Zhu*, Dingfan Chen*(co-first), Xiongfei Wu, Jiahui Geng, Zhuo Li, Jens Grossklags, Lei Ma
    Neural Information Processing Systems Datasets and Benchmarks Track (NeurIPS), 2024.
    PDF Code BibTex Slides Spotlight Presentation

  3. ECAI
    Inside the Black Box: Detecting Data Leakage in Pre-trained Language Encoders
    Yuan Xin, Zheng Li, Ning Yu, Dingfan Chen(corresponding), Mario Fritz, Michael Backes, Yang Zhang
    European Conference on Artificial Intelligence (ECAI), 2024.
    PDF BibTex

  4. NAACL
    PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
    Derui Zhu*, Dingfan Chen*(co-first), Qing Li, Zongxiong Chen, Lei Ma, Jens Grossklags, Mario Fritz
    Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), 2024.
    PDF Code BibTex

  5. TMLR
    A Unified View of Differentially Private Deep Generative Modeling
    Dingfan Chen, Raouf Kerkouche, Mario Fritz
    Transactions on Machine Learning Research (TMLR), 2024.
    PDF BibTex Survey Certification

  6. PoPETs
    FedLAP-DP: Federated Learning by Sharing Differentially Private Loss Approximations
    Hui-Po Wang, Dingfan Chen, Raouf Kerkouche, Mario Fritz
    Proceedings on Privacy Enhancing Technologies (PoPETs), 2024.
    PDF Code BibTex

  7. PoPETs
    Towards Biologically Plausible and Private Gene Expression Data Generation
    Dingfan Chen*, Marie Oestreich*, Tejumade Afonja*, Raouf Kerkouche, Matthias Becker, Mario Fritz
    Proceedings on Privacy Enhancing Technologies (PoPETs), 2024.
    PDF Code BibTex

  8. GCPR
    MargCTGAN: A “Marginally” Better CTGAN for the Low Sample Regime
    Tejumade Afonja, Dingfan Chen, Mario Fritz
    DAGM German Conference on Pattern Recognition (GCPR), 2023.
    PDF Code BibTex

  9. NeurIPS
    Private Set Generation with Discriminative Information
    Dingfan Chen, Raouf Kerkouche, Mario Fritz
    Neural Information Processing Systems (NeurIPS), 2022.
    PDF Code BibTex Poster Slides Video

  10. ICLR
    RelaxLoss: Defending Membership Inference Attacks without Losing Utility
    Dingfan Chen, Ning Yu, Mario Fritz
    International Conference on Learning Representations (ICLR), 2022.
    PDF Code BibTex Poster Slides Video Spotlight Presentation

  11. ICLR
    Responsible Disclosure of Generative Models Using Scalable Fingerprinting
    Ning Yu*, Vladislav Skripniuk*, Dingfan Chen, Larry Davis, Mario Fritz
    International Conference on Learning Representations (ICLR), 2022.
    PDF Code BibTex Poster Video Spotlight Presentation

  12. EXCLI Journal
    Privacy Considerations for Sharing Genomics Data
    Marie Oestreich, Dingfan Chen, Joachim L. Schultze, Mario Fritz, Matthias Becker
    EXCLI Journal, 2021.
    PDF BibTex

  13. ACSAC
    BadNL: Backdoor Attacks against NLP Models with Semantic-preserving Improvements
    Xiaoyi Chen, Ahmed Salem, Dingfan Chen, Michael Backes, Shiqing Ma, Qingni Shen, Zhonghai Wu, Yang Zhang
    Annual Computer Security Applications Conference (ACSAC), 2021
    PDF BibTex

  14. NeurIPS
    GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators
    Dingfan Chen, Tribhuvanesh Orekondy, Mario Fritz
    Neural Information Processing Systems (NeurIPS), 2020.
    PDF Code BibTex Poster Slides Video

  15. CCS
    Gan-leaks: A Taxonomy of Membership Inference Attacks against Generative Models
    Dingfan Chen, Ning Yu, Yang Zhang, Mario Fritz
    ACM Conference on Computer and Communications Security (CCS), 2020.
    PDF Code BibTex Poster Slides Video (short) Video (full)

Awards


Academic Activities

Reviewing


Teaching

Thesis Co-supervision