Code for the paper Exploring the zero-shot limit of FewRel. This repository implements a zero-shot relation extractor.
The dataset FewRel 1.0 has been created in the paper FewRel: A Large-Scale Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation and is available here.
An example relation extraction is in this notebook. The extractor needs a list of candidate relations in English
relations = ['noble title', 'founding date', 'occupation of a person']
extractor = RelationExtractor(model, tokenizer, relations)Then the model ranks the surface forms by the belief that the relation connects the entities in the text
extractor.rank(text='John Smith received an OBE', head='John Smith', tail='OBE')
[('noble title', 0.9690611883997917),
('occupation of a person', 0.0012609362602233887),
('founding date', 0.00024014711380004883)]This repository contains 4 training scripts related to the 4 models in the paper.
train_bert_large_with_squad.py
train_bert_large_without_squad.py
train_distillbert_with_squad.py
train_distillbert_without_squad.pyThere are also 4 scripts for validation
test_bert_large_with_squad.py
test_bert_large_without_squad.py
test_distillbert_with_squad.py
test_distillbert_without_squad.pyThe results as in the paper are
| Model | 0-shot 5-ways | 0-shot 10-ways |
|---|---|---|
| (1) Distillbert | 70.1±0.5 | 55.9±0.6 |
| (2) Bert Large | 80.8±0.4 | 69.6±0.5 |
| (3) Distillbert + SQUAD | 81.3±0.4 | 70.0±0.2 |
| (4) Bert Large + SQUAD | 86.0±0.6 | 76.2±0.4 |
@inproceedings{cetoli-2020-exploring,
title = "Exploring the zero-shot limit of {F}ew{R}el",
author = "Cetoli, Alberto",
booktitle = "Proceedings of the 28th International Conference on Computational Linguistics",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "International Committee on Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.coling-main.124",
doi = "10.18653/v1/2020.coling-main.124",
pages = "1447--1451",
abstract = "This paper proposes a general purpose relation extractor that uses Wikidata descriptions to represent the relation{'}s surface form. The results are tested on the FewRel 1.0 dataset, which provides an excellent framework for training and evaluating the proposed zero-shot learning system in English. This relation extractor architecture exploits the implicit knowledge of a language model through a question-answering approach.",
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent