Understanding and Detecting Hallucinations in Neural Machine Translation via Model Introspection

• Published in: Transactions of the Association for Computational Linguistics Vol. 11; pp. 546 - 564
• Main Authors: Xu, Weijia, Agrawal, Sweta, Briakou, Eleftheria, Martindale, Marianna J., Carpuat, Marine
• Format: Journal Article
• Language: English
• Published: One Broadway, 12th Floor, Cambridge, Massachusetts 02142, USA MIT Press 12.06.2023; MIT Press Journals, The; The MIT Press
• Subjects: Chinese languages; Elections; Employment; English language; German language; Hallucinations; Hypotheses; Linguistics; Machine translation; Translations

Neural sequence generation models are known to “hallucinate”, by producing outputs that are unrelated to the source text. These hallucinations are potentially harmful, yet it remains unclear in what conditions they arise and how to mitigate their impact. In this work, we first identify internal mode...