Chulin Xie

Contact: chulinx2@illinois.edu

xcl-2024-feb.jpg

Hi! I am a Ph.D. candidate in Computer Science at University of Illinois at Urbana-Champaign, advised by Prof. Bo Li.

I am broadly interested in trustworthy machine learning and optimization. I was a student researcher at Google Research, and research intern at Microsoft Research and NVIDIA Research.


News


Nov 3, 2024 RedCode agent benchmark is accepted to NeurIPS’24, and our work on memorization v.s. reasoning and crosslingual knowledge are in workshops.
May 3, 2024 Our work on DP synthetic text and Decoding Compressed Trust, got accepted to ICML 2024.
Feb 28, 2024 Our work on efficient FL personalization, PerAda and FedSelect, got accepted to CVPR 2024.
Jan 18, 2024 Our work on red-teaming tool for Diffusion Models and Hybrid FL got accepted to ICLR 2024.
Dec 12, 2023 Our LLM trustworthiness benchmark DecodingTrust won Outstanding Paper Award at NeurIPS!
Sep 3, 2023 Our work on differential privacy and certified robustness in FL got accepted to ACM CCS 2023.

Selected Work


  1. Differentially Private Synthetic Data via Foundation Model APIs 2: Text
    Chulin Xie, Zinan Lin, Arturs Backurs, Sivakanth Gopi, Da Yu, Huseyin A Inan, Harsha Nori, Haotian Jiang, Huishuai Zhang, Yin Tat Lee, Bo Li, and Sergey Yekhanin
    ICML 2024 (Spotlight)
  2. LLM-PBE: Assessing Data Privacy in Large Language Models
    Qinbin Li*, Junyuan Hong*, Chulin Xie*, Jeffrey Tan, Rachel Xin, Junyi Hou, Xavier Yin, Zhun Wang, Dan Hendrycks, Zhangyang Wang, Bo Li, Bingsheng He, and Dawn Song
    VLDB 2024 (Best Paper Nomination)
  3. PerAda: Parameter-Efficient Federated Learning Personalization with Generalization Guarantees
    Chulin Xie, De-An Huang, Wenda Chu, Daguang Xu, Chaowei Xiao, Bo Li, and Anima Anandkumar
    CVPR 2024
  4. Improving Privacy-Preserving Vertical Federated Learning by Efficient Communication with ADMM
    Chulin Xie, Pin-Yu Chen, Qinbin Li, Arash Nourian, Ce Zhang, and Bo Li
    SaTML 2024
  5. Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?
    Yu-Lin Tsai, Chia-Yi Hsu, Chulin Xie, Chih-Hsun Lin, Jia-You Chen, Bo Li, Pin-Yu Chen, Chia-Mu Yu, and Chun-Ying Huang
    ICLR 2024
  6. DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
    Boxin Wang*, Weixin Chen*, Hengzhi Pei*, Chulin Xie*, Mintong Kang*, Chenhui Zhang*, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T. Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, and Bo Li
    NeurIPS Datasets & Benchmarks 2023 (Oral) Outstanding Paper Award
  7. Unraveling the Connections between Privacy and Certified Robustness in Federated Learning Against Poisoning Attacks
    Chulin Xie, Yunhui Long, Pin-Yu Chen, Qinbin Li, Sanmi Koyejo, and Bo Li
    ACM CCS 2023
  8. CRFL: Certifiably Robust Federated Learning against Backdoor Attacks
    Chulin Xie, Minghao Chen, Pin-Yu Chen, and Bo Li
    ICML 2021 (Spotlight)

Teaching


  • Teaching Assistant for CS 446: Machine Learning, Fall 2023