UKP-Athene: Multi-Sentence Textual Entailment for Claim Verification

TL;DR

This work addresses scalable fact-checking with the FEVER dataset by integrating three components: entity-linking–driven document retrieval, an ESIM-based sentence ranking model for evidence selection, and an extended ESIM architecture that aggregates multiple sentences to perform multi-sentence textual entailment. The system demonstrates strong performance, achieving third place among 23 teams and substantial improvements over baselines across document retrieval, sentence selection, and entailment, with a FEVER score of 64.74. The approach emphasizes leveraging constituency-based entity extraction, Wikipedia-derived evidence, and attention-based fusion of evidence to support or refute claims. Public release of code and pretrained models facilitates replication and further research in automated fact verification and claim verification pipelines.

Abstract

The Fact Extraction and VERification (FEVER) shared task was launched to support the development of systems able to verify claims by extracting supporting or refuting facts from raw text. The shared task organizers provide a large-scale dataset for the consecutive steps involved in claim verification, in particular, document retrieval, fact extraction, and claim classification. In this paper, we present our claim verification pipeline approach, which, according to the preliminary results, scored third in the shared task, out of 23 competing systems. For the document retrieval, we implemented a new entity linking approach. In order to be able to rank candidate facts and classify a claim on the basis of several selected facts, we introduce two extensions to the Enhanced LSTM (ESIM).

Figures (2)

• Figure 1: Sentence selection model
• Figure 2: Extended ESIM for recognizing textual entailment