Automated Fact Extraction and Verification for Detecting False Information

Subramanian, Shyam

Etd

Automated Fact Extraction and Verification for Detecting False Information

Pubblico

Automated fact extraction and verification is a challenging task that involves finding relevant evidence sentences from a reliable corpus to verify the truthfulness of a claim. Existing models either (i) concatenate all the evidence sentences, leading to the inclusion of unnecessary sentences containing redundant, distracting, noisy or irrelevant information; or (ii) process each claim-evidence sentence pair separately and aggregate all of them later, missing the early combination of related sentences for more accurate claim verification. Unlike the prior works, in this thesis, we propose Hierarchical Evidence Set Modeling (HESM), a framework to extract evidence sets (each of which may contain multiple evidence sentences) and verify a claim to be SUPPORTED, REFUTED, or NOT ENOUGH INFO, by encoding and attending the claim-evidence set pairs at different levels of hierarchy. Each evidence set combines only the related sentences while limiting unnecessary sentences. Thus, our HESM framework overcomes the limitations of existing models that concatenates evidence sentences or aggregates individual claim-evidence sentence pairs. HESM consists of document retriever, multi-hop evidence retriever, and claim verification components. In the framework, we extract multiple evidence sets, and process and evaluate a claim based on each evidence set. Then, we aggregate all the evidence sets using word-level and evidence set-level attention for final verification of the claim. Our experimental results show that HESM outperforms 7 state-of-the-art methods for fact extraction and claim verification.

Creator