BEVERS: A General, Simple, and Performant Framework for Automatic Fact Verification

TL;DR

BEVERS is a simple yet highly effective framework for automatic fact verification that achieves state-of-the-art results on multiple datasets by optimizing each component of the pipeline.

Contribution

The paper introduces BEVERS, a tuned baseline system that attains top performance on FEVER and Scifact datasets through careful component optimization.

Findings

Achieves highest FEVER score and label accuracy.

Attains highest label accuracy on Scifact dataset.

Demonstrates effectiveness of optimized pipeline components.

Abstract

Automatic fact verification has become an increasingly popular topic in recent years and among datasets the Fact Extraction and VERification (FEVER) dataset is one of the most popular. In this work we present BEVERS, a tuned baseline system for the FEVER dataset. Our pipeline uses standard approaches for document retrieval, sentence selection, and final claim classification, however, we spend considerable effort ensuring optimal performance for each component. The results are that BEVERS achieves the highest FEVER score and label accuracy among all systems, published or unpublished. We also apply this pipeline to another fact verification dataset, Scifact, and achieve the highest label accuracy among all systems on that dataset as well. We also make our full code available.