@inproceedings{sundararajan-etal-2022-error,
    title = "Error Analysis of {T}o{TT}o Table-to-Text Neural {NLG} Models",
    author = "Sundararajan, Barkavi  and
      Sripada, Somayajulu  and
      Reiter, Ehud",
    editor = "Bosselut, Antoine  and
      Chandu, Khyathi  and
      Dhole, Kaustubh  and
      Gangal, Varun  and
      Gehrmann, Sebastian  and
      Jernite, Yacine  and
      Novikova, Jekaterina  and
      Perez-Beltrachini, Laura",
    booktitle = "Proceedings of the Second Workshop on Natural Language Generation, Evaluation, and Metrics (GEM)",
    month = dec,
    year = "2022",
    address = "Abu Dhabi, United Arab Emirates (Hybrid)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/add-orcids-2024-eacl/2022.gem-1.43/",
    doi = "10.18653/v1/2022.gem-1.43",
    pages = "456--470",
    abstract = "We report error analysis of outputs from four Table-to-Text generation models fine-tuned on ToTTo, an open-domain English language dataset. We carried out a manual error annotation of a subset of outputs (a total of 3,016 sentences) belonging to the topic of \textit{Politics} generated by these four models. Our error annotation focused on eight categories of errors. The error analysis shows that more than 46{\%} of sentences from each of the four models have been error-free. It uncovered some of the specific classes of errors; for example, \textit{WORD} errors (mostly verbs and prepositions) are the dominant errors in all four models and are the most complex ones among other errors. \textit{NAME} (mostly nouns) and \textit{NUMBER} errors are slightly higher in two of the \textit{GeM} benchmark models, whereas \textit{DATE-DIMENSION} and \textit{OTHER} categories of errors are more common in our Table-to-Text model. This in-depth error analysis is currently guiding us in improving our Table-to-Text model."
}