@inproceedings{park-etal-2026-pac,
    title = "{PAC}-{BENCH}: Evaluating Multi-Agent Collaboration under Privacy Constraints",
    author = "Park, Minjun  and
      Kim, Donghyun  and
      Ju, Hyeonjong  and
      Lim, Seungwon  and
      Choi, Dongwook  and
      Kwon, Taeyoon  and
      Kim, Minju  and
      Yeo, Jinyoung",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1552/",
    pages = "31030--31056",
    ISBN = "979-8-89176-395-1",
    abstract = "We are entering an era in which individuals and organizations increasingly deploy dedicated AI agents that interact and collaborate with other agents.However, the dynamics of multi-agent collaboration under privacy constraints remain poorly understood.In this work, we present $PAC\text{-}Bench$, a benchmark for systematic evaluation of multi-agent collaboration under privacy constraints.Experiments on $PAC\text{-}Bench$ show that privacy constraints substantially degrade collaboration performance and make outcomes depend more on the initiating agent than the partner.Further analysis reveals that this degradation is driven by recurring coordination breakdowns, including early-stage privacy violations, overly conservative abstraction, and privacy-induced hallucinations.Together, our findings identify privacy-aware multi-agent collaboration as a distinct and unresolved challenge that requires new coordination mechanisms beyond existing agent capabilities."
}