@inproceedings{tang-etal-2025-enhancing,
    title = "Enhancing Chain-of-Thought Reasoning via Neuron Activation Differential Analysis",
    author = "Tang, Yiru  and
      Zhou, Kun  and
      Min, Yingqian  and
      Zhao, Xin  and
      Sha, Jing  and
      Sheng, Zhichao  and
      Wang, Shijin",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.817/",
    pages = "16162--16170",
    ISBN = "979-8-89176-332-6",
    abstract = "Despite the impressive chain-of-thought(CoT) reasoning ability of large language models (LLMs), its underlying mechanisms remains unclear. In this paper, we explore the inner workings of LLM{'}s CoT ability via the lens of neurons in the feed-forward layers. We propose an efficient method to identify reasoning-critical neurons by analyzing their activation patterns under reasoning chains of varying quality. Based on it, we devise a rather simple intervention method that directly stimulates these reasoning-critical neurons, to guide the generation of high-quality reasoning chains. Extended experiments validate the effectiveness of our method and demonstrate the critical role these identified neurons play in CoT reasoning."
}