Skip to main navigation Skip to search Skip to main content
Kaidi XU

Prof. Kaidi XU, 許凱第

  • LI-6388

20172027

Research activity per year

Personal profile

Author IDs

ORCID iD: 0000-0003-4437-0671
Scopus Author ID: 57205687721

Impact

Metrics for Academic and Research Impact

Biography

Prof. XU has obtained his PhD degree at the Northeastern University, and he holds a MSc degree and a BSc degree from the Department of Computer Science at the University of Florida and Sichuan University respectively. His primary research focuses on Trustworthy AI, including formal verification, practical adversarial attacks, and uncertainty quantification. As a Principal Investigator, Prof. Xu has led multiple research projects funded by the U.S. NSF and national laboratories. His work bridges rigorous theoretical guarantees with practical AI robustness solutions, advancing the safety and reliability of AI systems in real-world applications.

Prior to joining us, Prof. XU was an Assistant Professor in the Department of Computer Science at Drexel University from 2021 to 2025. With extensively publication in top-tier venues across machine learning, computer vision, and natural language processing; and serving as a program committee member and area chair for leading conferences. In addition, Prof. Xu is a four-time champion of the International Verification of Neural Networks Competition (VNN-COMP 2021, 2022, 2023, 2024), a recipient of Drexel’s Faculty Research Excellence Award, and a winner of the Best Paper Award at the NeurIPS 2024 GenAI4Health workshop.

Research Interests/Areas

Trustworthy AI, Uncertainty Quantification, Formal Verification, AI in Health, Foundation Models

Expertise related to UN Sustainable Development Goals

In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This person’s work contributes towards the following SDG(s):

  1. SDG 3 - Good Health and Well-being
    SDG 3 Good Health and Well-being
  2. SDG 4 - Quality Education
    SDG 4 Quality Education
  3. SDG 7 - Affordable and Clean Energy
    SDG 7 Affordable and Clean Energy

Fingerprint

Dive into the research topics where Kaidi XU is active. These topic labels come from the works of this person. Together they form a unique fingerprint.
  • 1 Similar Profiles
View full fingerprint

Collaborations from the last five years

Recent external collaboration based on locations. Dive into details by clicking on the dots or
Explore network further
  • 55 RGC 32 - Refereed conference paper (with host publication)
  • 8 RGC 21 - Publication in refereed journal

  • Can Multimodal LLMs ‘See’ Science Instruction? Benchmarking Pedagogical Reasoning in K–12 Classroom Videos

    Shen, Y., He, P., Liu, H., Fan, J., Ji, Y., Li, T., Chen, T., Xu, K. & Liu, F., 2027, Artificial Intelligence in Education - 27th International Conference, AIED 2026, Proceedings - 27th International Conference, AIED 2026, Proceedings, Part II. Blanchard, E. G., Chen, G., Chi, M. & Isotani, S. (eds.). Springer, Cham, p. 414-423 10 p. (Lecture Notes in Computer Science; vol. 16582 LNAI).

    Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

  • Dialogue is Better Than Monologue: Instructing Medical LLMs via Strategical Conversation

    Liu, Z., Zhao, X., Peng, J., Duan, J., Zhu, Z., Chen, Q., Xu, K., Hu, X. & Chen, T., Mar 2026, Findings of the Association for Computational Linguistics: EACL 2026. Demberg, V., Inui, K. & Marquez, L. (eds.). Association for Computational Linguistics, p. 2858-2872 15 p. (Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL).

    Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

    Open Access
    File
    1 Downloads (CityUHK Scholars)

  • Safety Alignment of Large Language Models via Contrasting Safe and Harmful Distributions

    Zhang, X. (Co-first Author), Zhao, Z. (Co-first Author), Shi, W., Xu, K., Huang, D. & Hu, X., 2026, Proceedings of the 40th Annual AAAI Conference on Artificial Intelligence. AAAI Press, p. 34827-34835 (Proceedings of the AAAI Conference on Artificial Intelligence; vol. 40, no. 41).

    Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

    Open Access

  • DiffZOO: A Purely Query-Based Black-Box Attack for Red-teaming Text-to-Image Generative Model via Zeroth Order Optimization

    Dang, P., Hu, X., Li, D., Zhang, R., Guo, Q. & Xu, K., Apr 2025, NAACL 2025 - Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Proceedings of the Conference Findings. ACL Anthology, p. 17-31 (Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Proceedings of the Conference Findings, NAACL).

    Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

    Open Access
    File
    4   Link opens in a new tab Citations (Scopus)
    2 Downloads (CityUHK Scholars)

  • DynaCode: A Dynamic Complexity-Aware Code Benchmark for Evaluating Large Language Models in Code Generation

    Hu, W., Duan, J., Wei, C., Zhang, L., Zhang, Y. & Xu, K., Jul 2025, The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025): Findings of the Association for Computational Linguistics: ACL 2025. Che, W., Nabende, J., Shutova, E. & Pilehvar, M. T. (eds.). ACL Anthology, p. 21980-21997 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

    Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

    Open Access
    File
    2   Link opens in a new tab Citations (Scopus)
    3 Downloads (CityUHK Scholars)
View all 63 research outputs