@inproceedings{hashimoto-etal-2024-beam,
    title = "How Does Beam Search improve Span-Level Confidence Estimation in Generative Sequence Labeling?",
    author = "Hashimoto, Kazuma  and
      Naim, Iftekhar  and
      Raman, Karthik",
    editor = {V{\'a}zquez, Ra{\'u}l  and
      Celikkanat, Hande  and
      Ulmer, Dennis  and
      Tiedemann, J{\"o}rg  and
      Swayamdipta, Swabha  and
      Aziz, Wilker  and
      Plank, Barbara  and
      Baan, Joris  and
      de Marneffe, Marie-Catherine},
    booktitle = "Proceedings of the 1st Workshop on Uncertainty-Aware NLP (UncertaiNLP 2024)",
    month = mar,
    year = "2024",
    address = "St Julians, Malta",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/add-emnlp-2024-awards/2024.uncertainlp-1.6/",
    pages = "62--69",
    abstract = "Sequence labeling is a core task in text understanding for IE/IR systems. Text generation models have increasingly become the go-to solution for such tasks (e.g., entity extraction and dialog slot filling). While most research has focused on the labeling accuracy, a key aspect {--} of vital practical importance {--} has slipped through the cracks: understanding model confidence. More specifically, we lack a principled understanding of how to reliably gauge the confidence of a model in its predictions for each labeled span. This paper aims to provide some empirical insights on estimating model confidence for generative sequence labeling. Most notably, we find that simply using the decoder`s output probabilities \textbf{is not} the best in realizing well-calibrated confidence estimates. As verified over six public datasets of different tasks, we show that our proposed approach {--} which leverages statistics from top-$k$ predictions by a beam search {--} significantly reduces calibration errors of the predictions of a generative sequence labeling model."
}