Extractive and Abstractive Explanations for Fact-Checking and Evaluation of News
Ashkan Kazemi, Zehua Li, Ver\'onica P\'erez-Rosas, Rada Mihalcea
TL;DR
This paper investigates methods for generating natural language explanations for news claims to aid fact-checking, comparing extractive and abstractive approaches on misinformation datasets, with extractive methods showing promising results.
Contribution
It introduces and compares extractive and abstractive explanation methods for fact-checking, highlighting the effectiveness of an unsupervised extractive approach.
Findings
Extractive method outperforms abstractive in evaluations.
Biased TextRank effectively extracts relevant content.
Approaches tested on political and health news datasets.
Abstract
In this paper, we explore the construction of natural language explanations for news claims, with the goal of assisting fact-checking and news evaluation applications. We experiment with two methods: (1) an extractive method based on Biased TextRank -- a resource-effective unsupervised graph-based algorithm for content extraction; and (2) an abstractive method based on the GPT-2 language model. We perform comparative evaluations on two misinformation datasets in the political and health news domains, and find that the extractive method shows the most promise.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
MethodsMulti-Head Attention · Attention Is All You Need · Linear Layer · Refunds@Expedia|||How do I get a full refund from Expedia? · Cosine Annealing · Residual Connection · Softmax · Attention Dropout · Linear Warmup With Cosine Annealing · Layer Normalization