@inproceedings{ou-etal-2026-browseconf,
    title = "{B}rowse{C}onf: Confidence-Guided Test-Time Scaling for Web Agents",
    author = "Ou, Litu  and
      Li, Kuan  and
      Yin, Huifeng  and
      Zhang, Liwen  and
      Zhang, Zhongwang  and
      Wu, Xixi  and
      Ye, Rui  and
      Qiao, Zile  and
      Jiang, Yong  and
      Xie, Pengjun  and
      Huang, Fei  and
      Zhou, Jingren",
    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.21/",
    pages = "446--465",
    ISBN = "979-8-89176-395-1",
    abstract = "Confidence in LLMs is a useful indicator of model uncertainty and answer reliability. Existing work mainly focused on single-turn scenarios, while research on confidence in complex multi-turn interactions is limited. In this paper, we investigate whether LLM-based search agents have the ability to communicate their own confidence through verbalized confidence scores after long sequences of actions, a significantly more challenging task compared to outputting confidence in a single interaction. Experimenting on open-source agentic models, we first find that models exhibit much higher task accuracy at high confidence while having near-zero accuracy when confidence is low. Based on this observation, we propose Test-Time Scaling (TTS) methods that use confidence scores to determine answer quality, encourage the model to try again until reaching a satisfactory confidence level. Results show that our proposed methods significantly reduce token consumption while demonstrating competitive performance compared to baseline fixed budget TTS methods."
}