Few Shot Learning for Information Verification

TL;DR

This paper proposes a few-shot learning approach for information verification using pretrained language models and graph-based attention to handle hierarchical evidence, requiring minimal additional training.

Contribution

It introduces a novel method combining pretrained models with graph-based attention for fact verification with limited training data.

Findings

Effective verification with minimal training data

Utilizes hierarchical evidence from Wikipedia articles

Combines XLNET with graph attention mechanisms

Abstract

Information verification is quite a challenging task, this is because many times verifying a claim can require picking pieces of information from multiple pieces of evidence which can have a hierarchy of complex semantic relations. Previously a lot of researchers have mainly focused on simply concatenating multiple evidence sentences to accept or reject claims. These approaches are limited as evidence can contain hierarchical information and dependencies. In this research, we aim to verify facts based on evidence selected from a list of articles taken from Wikipedia. Pretrained language models such as XLNET are used to generate meaningful representations and graph-based attention and convolutions are used in such a way that the system requires little additional training to learn to verify facts.