Publications

2025

  1. promptguard_overview.jpg
    PromptGuard: Soft Prompt-Guided Unsafe Content Moderation for Text-to-Image Models
    Lingzhi Yuan, Xinfeng Li, Chejian Xu, and 6 more authors
    2025
    Arxiv Pre-print
  2. mmdt_overview.png
    MMDT: Decoding the Trustworthiness and Safety of Multimodal Foundation Models
    Chejian Xu, Jiawei Zhang, Zhaorun Chen, and 22 more authors
    2025
    ICLR 2025