Publications

Preprints

2026

  1. arXiv
    Fail-Closed Alignment for Large Language Models
    Zachary Coalson, Beth Sohler, Aiden Gabriel, and Sanghyun Hong
    2026
    arXiv Code
  2. arXiv
    Asking Forever: Universal Activations Behind Turn Amplification in Conversational LLMs
    Zachary Coalson, Bo Fang, and Sanghyun Hong
    2026
    arXiv
  3. arXiv
    Discovering Universal Activation Directions for PII Leakage in Language Models
    Leo Marchyok, Zachary Coalson, Sungho Keum, Sooel Son, and Sanghyun Hong
    2026
    arXiv

2025

  1. arXiv
    PrisonBreak: Jailbreaking Large Language Models with at Most Twenty-Five Targeted Bit-flips
    Zachary Coalson, Jeonghyun Woo, Chris S. Lin, Joyce Qu, Yu Sun, Shiyang Chen, Lishan Yang, Gururaj Saileshwar, Prashant Nair, Bo Fang, and Sanghyun Hong
    2025
    arXiv

Conference Papers

2025

  1. NeurIPS
    IF-Guide: Influence Function-Guided Detoxification of LLMs
    Zachary Coalson, Juhan Bae, Nicholas Carlini, and Sanghyun Hong
    In The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025
    arXiv Code
  2. SC
    Demystifying the Resilience of Large Language Model Inference: An End-to-End Perspective
    Yu Sun, Zachary Coalson, Shiyang Chen, Hang Liu, Zhao Zhang, Sanghyun Hong, Bo Fang, and Lishan Yang
    In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, 2025
    DOI Code
  3. ICCV
    Harnessing Input-Adaptive Inference for Efficient VLN
    Dongwoo Kang, Akhil Perincherry, Zachary Coalson, Aiden Gabriel, Stefan Lee, and Sanghyun Hong
    In International Conference on Computer Vision, 2025
    arXiv Code

2023

  1. NeurIPS
    BERT Lost Patience Won’t Be Robust to Adversarial Slowdown
    Zachary Coalson, Gabriel Ritter, Rakesh B Bobba, and Sanghyun Hong
    In Thirty-seventh Conference on Neural Information Processing Systems, 2023
    arXiv Code

Journal Articles

2025

  1. TMLR
    Hard Work Does Not Always Pay Off: On the Robustness of NAS to Data Poisoning
    Zachary Coalson, Huazheng Wang, Qingyun Wu, and Sanghyun Hong
    Transactions on Machine Learning Research, 2025
    arXiv