RelationPrompt: Leveraging Prompts to Generate Synthetic Data for Zero-Shot Relation Triplet Extraction (2022-03-17T00:00:00.000000Z)

TL;DR

This work unify language model prompts and structured text approaches to design a structured prompt template for generating synthetic relation samples when conditioning on relation label prompts (RelationPrompt), and designs a novel Triplet Search Decoding method.

Abstract

Despite the importance of relation extraction in building and representing knowledge, less research is focused on generalizing to unseen relations types. We introduce the task setting of Zero-Shot Relation Triplet Extraction (ZeroRTE) to encourage further research in low-resource relation extraction methods. Given an input sentence, each extracted triplet consists of the head entity, relation label, and tail entity where the relation label is not seen at the training stage. To solve ZeroRTE, we propose to synthesize relation examples by prompting language models to generate structured texts. Concretely, we unify language model prompts and structured text approaches to design a structured prompt template for generating synthetic relation samples when conditioning on relation label prompts (RelationPrompt). To overcome the limitation for extracting multiple relation triplets in a sentence, we design a novel Triplet Search Decoding method. Experiments on FewRel and Wiki-ZSL datasets show the efficacy of RelationPrompt for the ZeroRTE task and zero-shot relation classification. Our code and data are available at github.com/declare-lab/RelationPrompt.

Authors

Soujanya Poria

37 papers

Lidong Bing

19 papers

Luo Si

13 papers

TL;DR

Abstract

Authors

References41 items

ZeroGen: Efficient Zero-shot Learning via Dataset Generation

Generating Training Data with Language Models: Towards Zero-Shot Language Understanding

Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing

Learning Span-Level Interactions for Aspect Sentiment Triplet Extraction

Template-Based Named Entity Recognition Using BART

A Frustratingly Easy Approach for Entity and Relation Extraction

ZS-BERT: Towards Zero-Shot Relation Extraction with Attribute Representation Learning

Calibrate Before Use: Improving Few-Shot Performance of Language Models

Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm

Neural Data Augmentation via Example Extrapolation

Structured Prediction as Translation between Augmented Natural Languages

Two Are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders

DARE: Data Augmented Relation Extraction with GPT-2

How Much Knowledge Can You Pack into the Parameters of a Language Model?

Zero-shot Text Classification With Generative Language Models

Effective Modeling of Encoder-Decoder Architecture for Joint Entity and Relation Extraction

BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension

CTRL: A Conditional Transformer Language Model for Controllable Generation

Language Models as Knowledge Bases?

Enriching Pre-trained Language Model with Entity Information for Relation Classification

Entity-Relation Extraction as Multi-Turn Question Answering

Unsupervised Data Augmentation for Consistency Training

EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks

A Survey of Zero-Shot Learning

FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation

Contextual Augmentation: Data Augmentation by Words with Paradigmatic Relations

Train Once, Test Anywhere: Zero-Shot Learning for Text Classification

Decoupled Weight Decay Regularization

Zero-Shot Transfer Learning for Event Extraction

Zero-Shot Relation Extraction via Reading Comprehension

Distant Supervision for Relation Extraction with Sentence-Level Attention and Entity Descriptions

Question Answering on Freebase via Relation Extraction and Textual Evidence

Reasoning about Entailment with Neural Attention

Distilling the Knowledge in a Neural Network

Learning Entity and Relation Embeddings for Knowledge Graph Completion

A Neural Probabilistic Language Model

Leveraging Type Descriptions for Zero-shot Named Entity Recognition and Classification

Meta-tuning Language Models to Answer Prompts Better

DAGA: Data augmentation with a generation approach for low-resource tagging tasks

Language Models are Unsupervised Multitask Learners

Zero-shot Relation Classification as Textual Entailment

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL