@inproceedings{wang-etal-2025-enhancing-safe,
    title = "Enhancing Safe and Controllable Protein Generation via Knowledge Preference Optimization",
    author = "Wang, Yuhao  and
      Ding, Keyan  and
      Feng, Kehua  and
      Wang, Zeyuan  and
      Qin, Ming  and
      Li, Xiaotong  and
      Zhang, Qiang  and
      Chen, Huajun",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.616/",
    pages = "12553--12569",
    ISBN = "979-8-89176-251-0",
    abstract = "Protein language models have emerged as powerful tools for sequence generation, offering substantial advantages in functional optimization and *denovo* design. However, these models also present significant risks of generating harmful protein sequences, such as those that enhance viral transmissibility or evade immune responses. These concerns underscore critical biosafety and ethical challenges. To address these issues, we propose a Knowledge-guided Preference Optimization (KPO) framework that integrates prior knowledge via a Protein Safety Knowledge Graph. This framework utilizes an efficient graph pruning strategy to identify preferred sequences and employs reinforcement learning to minimize the risk of generating harmful proteins. Experimental results demonstrate that KPO effectively reduces the likelihood of producing hazardous sequences while maintaining high functionality, offering a robust safety assurance framework for applying generative models in biotechnology."
}