Automatic Stance Detection Using End-to-End Memory Networks

TL;DR

This paper introduces an end-to-end memory network for stance detection that jointly classifies stance and extracts evidence snippets, achieving state-of-the-art results on a benchmark dataset.

Contribution

It proposes a novel integrated neural architecture combining convolutional, recurrent, and similarity components for stance detection and evidence extraction.

Findings

Achieved state-of-the-art performance on Fake News Challenge dataset

Effectively jointly predicts stance and extracts evidence snippets

Demonstrates the effectiveness of end-to-end memory networks in stance detection

Abstract

We present a novel end-to-end memory network for stance detection, which jointly (i) predicts whether a document agrees, disagrees, discusses or is unrelated with respect to a given target claim, and also (ii) extracts snippets of evidence for that prediction. The network operates at the paragraph level and integrates convolutional and recurrent neural networks, as well as a similarity matrix as part of the overall architecture. The experimental evaluation on the Fake News Challenge dataset shows state-of-the-art performance.