@inproceedings{li-etal-2025-1,
    title = "1-2-3 Check: Enhancing Contextual Privacy in {LLM} via Multi-Agent Reasoning",
    author = "Li, Wenkai  and
      Sun, Liwen  and
      Guan, Zhenxiang  and
      Zhou, Xuhui  and
      Sap, Maarten",
    editor = "Novikova, Jekaterina",
    booktitle = "Proceedings of the The First Workshop on LLM Security (LLMSEC)",
    month = aug,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/corrections-2025-08/2025.llmsec-1.9/",
    pages = "115--128",
    ISBN = "979-8-89176-279-4",
    abstract = "Addressing contextual privacy concerns remains challenging in interactive settings where large language models (LLMs) process information from multiple sources. Building on the theory of contextual integrity, we introduce a multi-agent framework that decomposes privacy reasoning into specialized subtasks{---}extraction, classification{---}reducing the information load on any single agent while enabling iterative validation and more reliable adherence to contextual privacy norms. Experiments on the ConfAIde benchmark with two LLMs (GPT-4, Llama3) demonstrate that our multi-agent system substantially reduces private information leakage (36{\%} reduction) while maintaining the fidelity of public content compared to a single-agent system, showing the promise of multi-agent frameworks towards contextual privacy with LLMs."
}