EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Revisiting Meta-evaluation for Grammatical Error Correction

Masamune Kobayashi, Masato Mita, Mamoru Komachi

Transactions of the Association for Computational Linguistics · 15.07.2024 · 12 · 837-855

DOI: 10.1162/tacl_a_00676 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 15.07.2024. Verfasst von Masamune Kobayashi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Metrics are the foundation for automatic evaluation in grammatical error correction (GEC), with their evaluation of the metrics (meta-evaluation) relying on their correlation with human judgments. However, conventional meta-evaluations in English GEC encounter several challenges, including biases caused by inconsistencies in evaluation granularity and an outdated setup using classical systems. These problems can lead to misinterpretation of metrics and potentially hinder the applicability of GEC techniques. To address these issues, this paper proposes SEEDA, a new dataset for GEC meta-evaluation. SEEDA consists of corrections with human ratings along two different granularities: edit-based and sentence-based, covering 12 state-of-the-art systems including large language models, and two human corrections with different focuses. The results of improved correlations by aligning the granularity in the sentence-level meta-evaluation suggest that edit-based metrics may have been underestimated in existing studies. Furthermore, correlations of most metrics decrease when changing from classical to neural systems, indicating that traditional metrics are relatively poor at evaluating fluently corrected sentences with many edits.

Zitieren

APA

Kobayashi, M., Mita, M., & Komachi, M. (2024). Revisiting Meta-evaluation for Grammatical Error Correction. Transactions of the Association for Computational Linguistics, 12, 837-855. https://doi.org/10.1162/tacl_a_00676

Vancouver

Kobayashi M, Mita M, Komachi M. Revisiting Meta-evaluation for Grammatical Error Correction. Transactions of the Association for Computational Linguistics. 2024;12:837-855. doi:10.1162/tacl_a_00676

BibTeX
@article{Kobayashi2024,
  author = {Kobayashi, Masamune and Mita, Masato and Komachi, Mamoru},
  title = {Revisiting Meta-evaluation for Grammatical Error Correction},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2024},
  volume = {12},
  pages = {837--855},
  doi = {10.1162/tacl_a_00676}
}
Kostenlos starten →