@inproceedings{yu-etal-2026-thermometer,
    title = "Thermometer of Thoughts: Enhancing {LLM}{'}s Exploration via Attention Temperature Modulation",
    author = "Yu, Zhiyuan  and
      Xiao, Shijian  and
      Nguyen, Cam-Tu  and
      Yin, Zhangyue  and
      Xing, Lekai  and
      Li, Wenzhong  and
      Lu, Sanglu",
    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.200/",
    pages = "4355--4368",
    ISBN = "979-8-89176-390-6",
    abstract = "Improving the exploration of reasoning is essential for advancing Large Language Models' (LLMs) problem-solving performance. Current methods primarily rely on output-level stochasticity, which decode within fixed reasoning patterns of LLM and suffer from insufficient exploration. In this paper, we introduce adjusting attention temperature to directly modulate the model{'}s internal focus during reasoning, which enables a dynamic shift between exploratory and focused processing. We reveal that moderate adjustments preserve LLM{'}s reasoning capability while producing problem hardness-dependent benefits: higher temperatures facilitate solving complex tasks by encouraging wider exploration, whereas lower temperatures mitigate overthinking on simpler problems. Leveraging this insight, we propose a two-stage inference strategy: first, attention temperature scaling modulates the LLM{'}s reasoning patterns to diversify the reasoning traces; then, a difficulty-aware aggregation scheme is introduced to effectively identify the most reliable solution from the generated candidates. Extensive evaluations show that our method improves Pass@10 by 6.78{--}14.20{\%} and aggregation accuracy by 9.74{\%} across 7 reasoning benchmarks."
}