@inproceedings{wu-etal-2026-beyond,
    title = "Beyond Examples: Towards Automated Thought-level In-Context Reasoning for Large Language Models",
    author = "Wu, Jinyang  and
      Feng, Mingkuan  and
      Zhang, Shuai  and
      Che, Feihu  and
      Wen, Zhengqi  and
      Liao, Chonghua  and
      Yang, Ling  and
      Luo, Haoran  and
      Lian, Zheng  and
      Tao, Jianhua",
    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.135/",
    pages = "2955--2995",
    ISBN = "979-8-89176-390-6",
    abstract = "In-context learning (ICL) leverages demonstrations to enhance the performance of large language models (LLMs). However, traditional ICL struggles with complex reasoning mainly due to superficial, example-level implicit imitation. To address these limitations, we introduce **ThoughtICR**, an automated **Thought**-level **I**n-**C**ontext **R**easoning paradigm that shifts from surface-level examples to more guidance-oriented thought patterns. Specifically, we first define atomic reasoning actions and construct thought patterns on small-scale seed data using Monte Carlo Tree Search (MCTS). During inference, we dynamically select appropriate thought patterns based on target problem attributes, providing explicit guidance for model reasoning. Thanks to its automated and strategic design, our method enables seamless plug-and-play integration with various post-training techniques. Experimental results demonstrate that our method improves performance across different model sizes and generalizes effectively across reasoning domains. Using only small-scale seed data, we achieve 80.6{\%} accuracy on MATH and 62.5{\%} on AMC, surpassing GPT-4o{'}s 77.2{\%} and 57.5{\%}, respectively. Moreover, compared to test-time scaling methods, our approach reduces computational costs by over 10. Our code is available at https://github.com/jinyangwu/ThoughtICR."
}