search

declare-lab / RelationPrompt

This repository implements our ACL Findings 2022 research paper RelationPrompt: Leveraging Prompts to Generate Synthetic Data for Zero-Shot Relation Triplet Extraction. The goal of Zero-Shot Relation Triplet Extraction (ZeroRTE) is to extract relation triplets of the format (head entity, tail entity, relation), despite not having annotated data for the test relation labels.
MIT License
122 stars 16 forks source link

Unseen Labels #14

Open Cishenn opened 1 year ago

Cishenn commented 1 year ago

图片 请问为什么m增大, RelationPrompt会下降

Cishenn commented 1 year ago

Why does the RelationPrompt go down when m increases?

jbrry commented 1 year ago

m is the test set size, so as more labels are used for testing, the less there are for training. The training labels are used to generate synthetic data, so having more training labels should lead to better generalisation.