A Multi-Level Attention Model for Evidence-Based Fact Checking
This work addresses fact checking for claims using evidence from text, offering a competitive but incremental improvement over existing methods.
The paper tackles the problem of evidence-based fact checking by learning representations to verify claims against textual evidence, and presents a model that outperforms graph-based approaches with 1.09% and 1.42% improvements in label accuracy and FEVER score on the FEVER dataset.
Evidence-based fact checking aims to verify the truthfulness of a claim against evidence extracted from textual sources. Learning a representation that effectively captures relations between a claim and evidence can be challenging. Recent state-of-the-art approaches have developed increasingly sophisticated models based on graph structures. We present a simple model that can be trained on sequence structures. Our model enables inter-sentence attentions at different levels and can benefit from joint training. Results on a large-scale dataset for Fact Extraction and VERification (FEVER) show that our model outperforms the graph-based approaches and yields 1.09% and 1.42% improvements in label accuracy and FEVER score, respectively, over the best published model.