DialFact: A Benchmark for Fact-Checking in Dialogue
Preprint
Abstract
Fact-checking is an essential tool to mitigate the spread of misinformation and disinformation. We introduce the task of fact-checking in dialogue, which is a relatively unexplored area. We construct DialFact, a testing benchmark dataset of 22,245 annotated conversational claims, paired with pieces of evidence from Wikipedia. There are three sub-tasks in DialFact: 1) Verifiable claim detection task distinguishes whether a response carries verifiable factual information; 2) Evidence retrieval task retrieves the most relevant Wikipedia snippets as evidence; 3) Claim verification task predicts a dialogue response to be supported, refuted, or not enough information. We found that existing fact-checking models trained on non-dialogue data like FEVER fail to perform well on our task, and thus, we propose a simple yet data-efficient solution to effectively improve fact-checking performance in dialogue. We point out unique challenges in DialFact such as handling the colloquialisms, coreferences and retrieval ambiguities in the error analysis to shed light on future research in this direction.Keywords
All Related Versions
This publication has 35 references indexed in Scilit:
- A Survey on Automated Fact-CheckingTransactions of the Association for Computational Linguistics, 2022
- A Review on Fact Extraction and VerificationACM Computing Surveys, 2021
- WinoGrandeCommunications of the ACM, 2021
- HoVer: A Dataset for Many-Hop Fact Extraction And Claim Verification2020
- Generating Label Cohesive and Well-Formed Adversarial Claims2020
- Fact or Fiction: Verifying Scientific Claims2020
- Dense Passage Retrieval for Open-Domain Question Answering2020
- Coreferential Reasoning Learning for Language Representation2020
- DeSePtion: Dual Sequence Prediction and Adversarial Examples for Improved Fact-Checking2020
- Fine-grained Fact Verification with Kernel Graph Attention Network2020