Andreas Vlachos

Automated Fact-checking, an NLP perspective

Misinformation is considered one of the major challenges of our times resulting in numerous efforts against it. Fact-checking, the task of assessing whether a claim is true or false, is considered a key in reducing its impact. In this talk I will present our work on automating this task using natural language processing, moving beyond simply classifying claims as true or false in the following aspects: incorporating tabular information, neurosymbolic inference, going beyond English, and using a search engine as a source of evidence. I will conclude with a novel proposal on evaluating language models based on plausibility, which bridges the gap between factuality and creativity.
 

back to overview
 

Biography

Andreas Vlachos is a professor of Natural Language Processing and Machine Learning at the Department of Computer Science and Technology at the University of Cambridge and a Dinesh Dhamija fellow of Fitzwilliam College. Current projects include dialogue modelling, automated fact checking and imitation learning. Andreas has also worked on semantic parsing, natural language generation and summarization, language modelling, information extraction, active learning, clustering and biomedical text mining. His research team is supported by grants from ERC, EPSRC, ESRC, Facebook, Amazon, Google, Huawei, the Alan Turing Institute and the Isaac Newton Trust.
Footer

Imprint / Privacy