@inproceedings{kasprova-etal-2026-polite,
    title = "Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems",
    author = "Kasprova, Vira  and
      Parulekar, Amruta  and
      AlRabah, Abdulrahman  and
      Agaram, Krishna  and
      Garg, Ritwik  and
      Jha, Sagar  and
      Bozdag, Nimet Beyza  and
      Hakkani-Tur, Dilek",
    editor = "Choi, Jinho D.  and
      Chen, Yun-Nung  and
      Funakoshi, Kotaro  and
      Emami, Ali",
    booktitle = "Proceedings of the 27th Annual Meeting of the Special Interest Group on Discourse and Dialogue",
    month = aug,
    year = "2026",
    address = "Atlanta, Georgia, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/paragraph-normalization/2026.sigdial-1.56/",
    pages = "795--814",
    abstract = "Large language models (LLMs) often exhibit sycophancy: agreement with user stance even when it conflicts with the model{'}s opinion. While prior work has mostly studied this in single-agent settings, it remains underexplored in collaborative multi-agent systems. We ask whether awareness of other agents' sycophancy levels influences discussion outcomes. To investigate this, we run controlled experiments with six open-source LLMs, providing agents with peer sycophancy rankings that estimate each peer{'}s tendency toward sycophancy. These rankings are based on scores calculated using various static (pre-discussion) and dynamic (online) strategies. We find that providing sycophancy priors reduces the influence of sycophancy-prone peers, mitigates error-cascades, and improves final discussion accuracy by an absolute 10.5{\%}. Thus, this is a lightweight and efficient way to reduce model sycophancy during discussions and subsequently improve downstream accuracy."
}