Fact or Fiction? Improving Fact Verification with Knowledge Graphs through Simplified Subgraph Retrievals

TL;DR

This paper introduces simplified subgraph retrieval methods for fact verification using knowledge graphs, achieving higher accuracy with less computational resources on the FactKG dataset.

Contribution

It proposes a novel approach that simplifies evidence retrieval from knowledge graphs, improving efficiency and accuracy over existing methods.

Findings

Better test-set accuracy with simplified retrieval methods

Reduced computational resource requirements

Effective verification on FactKG dataset

Abstract

Despite recent success in natural language processing (NLP), fact verification still remains a difficult task. Due to misinformation spreading increasingly fast, attention has been directed towards automatically verifying the correctness of claims. In the domain of NLP, this is usually done by training supervised machine learning models to verify claims by utilizing evidence from trustworthy corpora. We present efficient methods for verifying claims on a dataset where the evidence is in the form of structured knowledge graphs. We use the FactKG dataset, which is constructed from the DBpedia knowledge graph extracted from Wikipedia. By simplifying the evidence retrieval process, from fine-tuned language models to simple logical retrievals, we are able to construct models that both require less computational resources and achieve better test-set accuracy.