作者位次 #6/9ACL 2026Poster· Natural Language Processing
Awakening Dormant Experts:Counterfactual Routing to Mitigate M o E Hallucinations
Wentao Hu · Yanbo Zhai · Xiaohui Hu · Mingkuan Zhao · Shanhong yu · Xue Liu · Kaidong Yu · Shuangyong Song · Xuelong Li
摘要
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 fac...