Few-Shot Document-Level Relation Extraction

Nicholas Popovič, Michael Färber

2022Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies15 citationsDOI Open Access PDF

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-ofthe-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 1 .

Topics & Concepts

Computer scienceBenchmark (surveying)Relationship extractionDomain adaptationSentenceSet (abstract data type)Relation (database)Domain (mathematical analysis)Artificial intelligenceCode (set theory)Construct (python library)Natural language processingAdaptation (eye)Training setShot (pellet)Information retrievalInformation extractionData miningChemistryOpticsGeodesyProgramming languageOrganic chemistryMathematicsPhysicsMathematical analysisClassifier (UML)GeographyTopic ModelingNatural Language Processing TechniquesText Readability and Simplification