Understanding and Detecting Hallucinations in Neural Machine Translation via Model Introspection

Open in new window