Highly Parallel Autoregressive Entity Linking with Discriminative Correction

Nicola De Cao, Wilker Aziz, Ivan Titov


Abstract
Generative approaches have been recently shown to be effective for both Entity Disambiguation and Entity Linking (i.e., joint mention detection and disambiguation). However, the previously proposed autoregressive formulation for EL suffers from i) high computational cost due to a complex (deep) decoder, ii) non-parallelizable decoding that scales with the source sequence length, and iii) the need for training on a large amount of data. In this work, we propose a very efficient approach that parallelizes autoregressive linking across all potential mentions and relies on a shallow and efficient decoder. Moreover, we augment the generative objective with an extra discriminative component, i.e., a correction term which lets us directly optimize the generator’s ranking. When taken together, these techniques tackle all the above issues: our model is >70 times faster and more accurate than the previous generative method, outperforming state-of-the-art approaches on the standard English dataset AIDA-CoNLL. Source code available at https://github.com/nicola-decao/efficient-autoregressive-EL
Anthology ID:
2021.emnlp-main.604
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7662–7669
Language:
URL:
https://aclanthology.org/2021.emnlp-main.604
DOI:
10.18653/v1/2021.emnlp-main.604
Bibkey:
Cite (ACL):
Nicola De Cao, Wilker Aziz, and Ivan Titov. 2021. Highly Parallel Autoregressive Entity Linking with Discriminative Correction. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7662–7669, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Highly Parallel Autoregressive Entity Linking with Discriminative Correction (De Cao et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.604.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.604.mp4
Code
 nicola-decao/efficient-autoregressive-EL
Data
AIDA CoNLL-YAGOCoNLL