@inproceedings{hu-etal-2026-awakening,
    title = "Awakening Dormant Experts:Counterfactual Routing to Mitigate {M}o{E} Hallucinations",
    author = "Hu, Wentao  and
      Zhai, Yanbo  and
      Hu, Xiaohui  and
      Zhao, Mingkuan  and
      yu, Shanhong  and
      Liu, Xue  and
      Yu, Kaidong  and
      Song, Shuangyong  and
      Li, Xuelong",
    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.2187/",
    pages = "47257--47271",
    ISBN = "979-8-89176-390-6",
    abstract = "Sparse Mixture-of-Experts (MoE) models have achieved remarkable scalability, yet they remain vulnerable to hallucinations, particularly when processing long-tail knowledge. We identify that this fragility stems from static Top-k routing: routers tend to favor high-frequency patterns over rare factual associations. Consequently, ``specialist experts'' possessing critical long-tail knowledge are often assigned low gating scores and remain ``dormant''{---}under-prioritized for specific tokens despite their proven causal importance on other inputs. To address this, we propose Counterfactual Routing (CoR), a training-free inference framework designed to awaken these dormant experts. CoR integrates layer-wise perturbation analysis with the Counterfactual Expert Impact (CEI) metric to dynamically shift computational resources from syntax-dominant to knowledge-intensive layers while maintaining a constant total activation count, effectively retrieving causally decisive experts via virtual ablation. Extensive experiments on TruthfulQA, FACTOR, and TriviaQA demonstrate that CoR improves factual accuracy by 3.1{\%} on average without increasing the inference budget, establishing a superior Pareto frontier compared to static scaling strategies."
}