EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?

Byung-Doh Oh, William Schuler

Transactions of the Association for Computational Linguistics · 27.03.2023 · 11 · 336-350

DOI: 10.1162/tacl_a_00548 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 27.03.2023. Verfasst von Byung-Doh Oh und 1 weiteren Autor:innen. Die Arbeit wurde bisher 93-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This work presents a linguistic analysis into why larger Transformer-based pre-trained language models with more parameters and lower perplexity nonetheless yield surprisal estimates that are less predictive of human reading times. First, regression analyses show a strictly monotonic, positive log-linear relationship between perplexity and fit to reading times for the more recently released five GPT-Neo variants and eight OPT variants on two separate datasets, replicating earlier results limited to just GPT-2 (Oh et al., 2022). Subsequently, analysis of residual errors reveals a systematic deviation of the larger variants, such as underpredicting reading times of named entities and making compensatory overpredictions for reading times of function words such as modals and conjunctions. These results suggest that the propensity of larger Transformer-based models to ‘memorize’ sequences during training makes their surprisal estimates diverge from humanlike expectations, which warrants caution in using pre-trained language models to study human language processing.

Zitieren

APA

Oh, B., & Schuler, W. (2023). Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?. Transactions of the Association for Computational Linguistics, 11, 336-350. https://doi.org/10.1162/tacl_a_00548

Vancouver

Oh B, Schuler W. Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?. Transactions of the Association for Computational Linguistics. 2023;11:336-350. doi:10.1162/tacl_a_00548

BibTeX
@article{Oh2023,
  author = {Oh, Byung-Doh and Schuler, William},
  title = {Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2023},
  volume = {11},
  pages = {336--350},
  doi = {10.1162/tacl_a_00548}
}
Kostenlos starten →