EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Efficient Long-Text Understanding with Short-Text Models

Maor Ivgi, Uri Shaham, Jonathan Berant

Transactions of the Association for Computational Linguistics · 22.03.2023 · 11 · 284-299

DOI: 10.1162/tacl_a_00547 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 22.03.2023. Verfasst von Maor Ivgi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transformer-based pretrained language models (LMs) are ubiquitous across natural language understanding, but cannot be applied to long sequences such as stories, scientific articles, and long documents due to their quadratic complexity. While a myriad of efficient transformer variants have been proposed, they are typically based on custom implementations that require expensive pretraining from scratch. In this work, we propose SLED: SLiding-Encoder and Decoder, a simple approach for processing long sequences that re-uses and leverages battle-tested short-text pretrained LMs. Specifically, we partition the input into overlapping chunks, encode each with a short-text LM encoder and use the pretrained decoder to fuse information across chunks (fusion-in-decoder). We illustrate through controlled experiments that SLED offers a viable strategy for long text understanding and evaluate our approach on SCROLLS, a benchmark with seven datasets across a wide range of language understanding tasks. We find that SLED is competitive with specialized models that are up to 50x larger and require a dedicated and expensive pretraining step.

Zitieren

APA

Ivgi, M., Shaham, U., & Berant, J. (2023). Efficient Long-Text Understanding with Short-Text Models. Transactions of the Association for Computational Linguistics, 11, 284-299. https://doi.org/10.1162/tacl_a_00547

Vancouver

Ivgi M, Shaham U, Berant J. Efficient Long-Text Understanding with Short-Text Models. Transactions of the Association for Computational Linguistics. 2023;11:284-299. doi:10.1162/tacl_a_00547

BibTeX
@article{Ivgi2023,
  author = {Ivgi, Maor and Shaham, Uri and Berant, Jonathan},
  title = {Efficient Long-Text Understanding with Short-Text Models},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2023},
  volume = {11},
  pages = {284--299},
  doi = {10.1162/tacl_a_00547}
}
Kostenlos starten →