Please refer to my Google Scholar for a complete publication list.


  1. Effective Backdoor Defense by Exploiting Sensitivity of Poisoned Samples (Spotlight)
    Weixin Chen, Baoyuan Wu, and Haoqian Wang
    NeurIPS 2022


  1. TrojDiff: Trojan Attacks on Diffusion Models with Diverse Targets
    Weixin Chen, Dawn Song, and Bo Li
    CVPR 2023
  2. DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models (Oral)
    Boxin Wang, Weixin Chen, Hengzhi Pei, and 8 more authors
    NeurIPS Datasets and Benchmarks 2023