EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Holmes ⌕ A Benchmark to Assess the Linguistic Competence of Language Models

Andreas Waldis, Andreas Waldis, Yotam Perlitz, Leshem Choshen, Leshem Choshen, Yufang Hou, Iryna Gurevych

Transactions of the Association for Computational Linguistics · 04.12.2024 · 12 · 1616-1647

DOI: 10.1162/tacl_a_00718 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 04.12.2024. Verfasst von Andreas Waldis und 6 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 140 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We introduce Holmes, a new benchmark designed to assess language models’ (LMs’) linguistic competence—their unconscious understanding of linguistic phenomena. Specifically, we use classifier-based probing to examine LMs’ internal representations regarding distinct linguistic phenomena (e.g., part-of-speech tagging). As a result, we meet recent calls to disentangle LMs’ linguistic competence from other cognitive abilities, such as following instructions in prompting-based evaluations. Composing Holmes, we review over 270 probing studies and include more than 200 datasets to assess syntax, morphology, semantics, reasoning, and discourse phenomena. Analyzing over 50 LMs reveals that, aligned with known trends, their linguistic competence correlates with model size. However, surprisingly, model architecture and instruction tuning also significantly influence performance, particularly in morphology and syntax. Finally, we propose FlashHolmes, a streamlined version that reduces the computation load while maintaining high-ranking precision.

Zitieren

APA

Waldis, A., Waldis, A., Perlitz, Y., Choshen, L., Choshen, L., Hou, Y., & Gurevych, I. (2024). Holmes ⌕ A Benchmark to Assess the Linguistic Competence of Language Models. Transactions of the Association for Computational Linguistics, 12, 1616-1647. https://doi.org/10.1162/tacl_a_00718

Vancouver

Waldis A, Waldis A, Perlitz Y, Choshen L, Choshen L, Hou Y, et al. Holmes ⌕ A Benchmark to Assess the Linguistic Competence of Language Models. Transactions of the Association for Computational Linguistics. 2024;12:1616-1647. doi:10.1162/tacl_a_00718

BibTeX
@article{Waldis2024,
  author = {Waldis, Andreas and Waldis, Andreas and Perlitz, Yotam and Choshen, Leshem and Choshen, Leshem and Hou, Yufang and Gurevych, Iryna},
  title = {Holmes ⌕ A Benchmark to Assess the Linguistic Competence of Language Models},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2024},
  volume = {12},
  pages = {1616--1647},
  doi = {10.1162/tacl_a_00718}
}
Kostenlos starten →