@inproceedings{rai-etal-2026-taro,
    title = "{TAR}o: Token-level Adaptive Routing for {LLM} Test-time Alignment",
    author = "Rai, Arushi  and
      Zhang, Qiang  and
      Zeng, Hanqing  and
      Zhang, Yunkai  and
      Tamboli, Dipesh  and
      Fan, Xiangjun  and
      Zhao, Zhuokai  and
      Zhang, Lizhu",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.50/",
    pages = "1004--1025",
    ISBN = "979-8-89176-395-1",
    abstract = "Large language models (LLMs) exhibit strong reasoning capabilities but typically require expensive post-training to reach high performance. Recent test-time alignment methods offer a lightweight alternative, but have been explored mainly for preference alignment rather than reasoning. To bridge this gap, we propose Token-level Adaptive Routing (TARo), which steers frozen LLMs toward structured reasoning entirely at inference time. Specifically, we first train reward models on step-wise mathematical traces to capture fine-grained logical consistency signals, then introduce a learnable token-level router that automatically controls the guidance of the reward model to the base model. Extensive experiments show that TARo significantly improves reasoning performance by up to +22.4{\%} over base model and +8.4{\%} over existing token-level test-time alignment methods, while also boosting out-of-distribution clinical reasoning (MedXpertQA) and instruction following (AlpacaEval). Furthermore, TARo also generalizes from small to large backbones without retraining, extending test-time alignment from preference optimization to robust, cross-domain reasoning."
}