Few-Shot Document-Level Relation Extraction

Nicholas Popovic, Michael Färber


Abstract
We present FREDo, a few-shot document-level relation extraction (FSDLRE) benchmark. As opposed to existing benchmarks which are built on sentence-level relation extraction corpora, we argue that document-level corpora provide more realism, particularly regarding none-of-the-above (NOTA) distributions. Therefore, we propose a set of FSDLRE tasks and construct a benchmark based on two existing supervised learning data sets, DocRED and sciERC. We adapt the state-of-the-art sentence-level method MNAV to the document-level and develop it further for improved domain adaptation. We find FSDLRE to be a challenging setting with interesting new characteristics such as the ability to sample NOTA instances from the support set. The data, code, and trained models are available online (https://github.com/nicpopovic/FREDo).
Anthology ID:
2022.naacl-main.421
Volume:
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Month:
July
Year:
2022
Address:
Seattle, United States
Editors:
Marine Carpuat, Marie-Catherine de Marneffe, Ivan Vladimir Meza Ruiz
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5733–5746
Language:
URL:
https://aclanthology.org/2022.naacl-main.421
DOI:
10.18653/v1/2022.naacl-main.421
Bibkey:
Cite (ACL):
Nicholas Popovic and Michael Färber. 2022. Few-Shot Document-Level Relation Extraction. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 5733–5746, Seattle, United States. Association for Computational Linguistics.
Cite (Informal):
Few-Shot Document-Level Relation Extraction (Popovic & Färber, NAACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.naacl-main.421.pdf
Video:
 https://aclanthology.org/2022.naacl-main.421.mp4
Code
 nicpopovic/fredo
Data
FREDoDocREDSciERC