@inproceedings{hou-etal-2026-pass,
    title = "{PAS}s-{M}o{E}: Mitigating Misaligned Co-drift among Router and Experts via Pathway Activation Subspaces for Continual Learning",
    author = "Hou, ZhiYan  and
      Guo, Haiyun  and
      Ma, Haokai  and
      Sun, Yandu  and
      Yang, Yonghui  and
      Wang, Jinqiao",
    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.1474/",
    pages = "31959--31972",
    ISBN = "979-8-89176-390-6",
    abstract = "Continual instruction tuning (CIT) requires multimodal large language models (MLLMs) to adapt to a stream of tasks without forgetting prior capabilities. A common strategy is to isolate updates by routing inputs to different LoRA experts. However, existing LoRA-based Mixture-of-Experts (MoE) methods often jointly update the router and experts in an indiscriminate way, causing the router{'}s preferences to co-drift with experts' adaptation pathways and gradually deviate from early-stage input{--}expert specialization. We term this as ***Misaligned Co-drift***, which blurs expert responsibilities and exacerbates forgetting. To address this, we introduce the ***pathway activation subspace (PASs)***, a LoRA-induced subspace that reflects which low-rank pathway directions an input activates in each expert, providing a capability-aligned coordinate system for routing and preservation. Based on PASs, we propose a fixed-capacity PASs-based MoE{--}LoRA method with two components: PAS-guided Reweighting, which calibrates routing using each expert{'}s pathway activation signals, and PAS-aware Rank Stabilization, which selectively stabilizes rank directions important to previous tasks. Experiments on a CIT benchmark show that our approach consistently outperforms a range of conventional continual learning baselines and MoE{--}LoRA variants in both accuracy and resistance to forgetting, without increasing model parameters. Our code is publicly available at \url{https://github.com/yueluoshuangtian/PASs-MoE}."
}