Towards Realistic Few-Shot Relation Extraction: A New Meta Dataset and Evaluation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Alam, Fahmida, Islam, Md Asiful, Vacareanu, Robert, Surdeanu, Mihai
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913302343319552
author Alam, Fahmida
Islam, Md Asiful
Vacareanu, Robert
Surdeanu, Mihai
author_facet Alam, Fahmida
Islam, Md Asiful
Vacareanu, Robert
Surdeanu, Mihai
contents We introduce a meta dataset for few-shot relation extraction, which includes two datasets derived from existing supervised relation extraction datasets NYT29 (Takanobu et al., 2019; Nayak and Ng, 2020) and WIKIDATA (Sorokin and Gurevych, 2017) as well as a few-shot form of the TACRED dataset (Sabo et al., 2021). Importantly, all these few-shot datasets were generated under realistic assumptions such as: the test relations are different from any relations a model might have seen before, limited training data, and a preponderance of candidate relation mentions that do not correspond to any of the relations of interest. Using this large resource, we conduct a comprehensive evaluation of six recent few-shot relation extraction methods, and observe that no method comes out as a clear winner. Further, the overall performance on this task is low, indicating substantial need for future research. We release all versions of the data, i.e., both supervised and few-shot, for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2404_04445
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Towards Realistic Few-Shot Relation Extraction: A New Meta Dataset and Evaluation
Alam, Fahmida
Islam, Md Asiful
Vacareanu, Robert
Surdeanu, Mihai
Computation and Language
Information Retrieval
We introduce a meta dataset for few-shot relation extraction, which includes two datasets derived from existing supervised relation extraction datasets NYT29 (Takanobu et al., 2019; Nayak and Ng, 2020) and WIKIDATA (Sorokin and Gurevych, 2017) as well as a few-shot form of the TACRED dataset (Sabo et al., 2021). Importantly, all these few-shot datasets were generated under realistic assumptions such as: the test relations are different from any relations a model might have seen before, limited training data, and a preponderance of candidate relation mentions that do not correspond to any of the relations of interest. Using this large resource, we conduct a comprehensive evaluation of six recent few-shot relation extraction methods, and observe that no method comes out as a clear winner. Further, the overall performance on this task is low, indicating substantial need for future research. We release all versions of the data, i.e., both supervised and few-shot, for future research.
title Towards Realistic Few-Shot Relation Extraction: A New Meta Dataset and Evaluation
topic Computation and Language
Information Retrieval
url https://arxiv.org/abs/2404.04445