Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation
Transactions of the Association for Computational Linguistics · 07.08.2023 · 11 · 941-959
DOI: 10.1162/tacl_a_00582 ↗Kurz gefasst
Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 07.08.2023. Verfasst von Fei Huang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Non-AutoRegressive (NAR) text generation models have drawn much attention because of their significantly faster decoding speed and good generation quality in machine translation. However, in a wider range of text generation tasks, existing NAR models lack proper pre-training, making them still far behind the pre-trained autoregressive models. In this paper, we propose Pre-trained Directed Acyclic Transformer (PreDAT) and a novel pre-training task to promote prediction consistency in NAR generation. Experiments on five text generation tasks show that our PreDAT remarkably outperforms existing pre-trained NAR models (+4.2 score on average) and even achieves better results than pre-trained autoregressive baselines in n-gram-based metrics, along with 17 times speedup in throughput. Further analysis shows that PreDAT benefits from the unbiased prediction order that alleviates the error accumulation problem in autoregressive generation, which provides new insights into the advantages of NAR generation.1
Zitieren
APA
Huang, F., Ke, P., & Huang, M. (2023). Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation. Transactions of the Association for Computational Linguistics, 11, 941-959. https://doi.org/10.1162/tacl_a_00582
Vancouver
Huang F, Ke P, Huang M. Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation. Transactions of the Association for Computational Linguistics. 2023;11:941-959. doi:10.1162/tacl_a_00582
BibTeX
@article{Huang2023,
author = {Huang, Fei and Ke, Pei and Huang, Minlie},
title = {Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation},
journal = {Transactions of the Association for Computational Linguistics},
year = {2023},
volume = {11},
pages = {941--959},
doi = {10.1162/tacl_a_00582}
}
EUVIMED