@inproceedings{zhang-etal-2026-dybbt,
    title = "{D}y{BBT}: Dynamic Balance via Bandit-inspired Targeting for Dialog Policy with Cognitive Dual Systems",
    author = "Zhang, Shuyu  and
      Wei, Yifan  and
      Yuan, Jialuo  and
      Wang, Xinru  and
      Zhu, Yanmin  and
      Liu, Yujie  and
      Li, Bin",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.2180/",
    pages = "47079--47111",
    ISBN = "979-8-89176-390-6",
    abstract = "Task oriented dialog systems often rely on static exploration strategies that do not adapt to dynamic dialog contexts, leading to inefficient exploration and suboptimal performance. We propose DyBBT, a novel dialog policy learning framework that formalizes the exploration challenge through a structured cognitive state space $\mathcal{C}$ that captures dialog progression, user uncertainty, and slot dependency. DyBBT proposes a bandit-inspired meta-controller that dynamically switches between a fast intuitive inference (System 1) and a slow deliberative reasoner (System 2) based on real-time cognitive states and visitation counts. Extensive experiments on single- and multi-domain benchmarks show that DyBBT achieves SOTA performance in success rate, efficiency, and generalization, with human evaluations confirming that its decisions are well-aligned with expert judgment."
}