Generation Challenges: Results of the Accuracy Evaluation Shared Task
This addresses the problem of ensuring reliable text generation for users in domains like sports reporting, but it is incremental as it builds on existing evaluation techniques.
The paper tackled the challenge of evaluating factual accuracy in neural NLG systems for sports reporting, finding that the best submissions performed well but all automatic methods struggled with complex errors like incorrect computation or inference.
The Shared Task on Evaluating Accuracy focused on techniques (both manual and automatic) for evaluating the factual accuracy of texts produced by neural NLG systems, in a sports-reporting domain. Four teams submitted evaluation techniques for this task, using very different approaches and techniques. The best-performing submissions did encouragingly well at this difficult task. However, all automatic submissions struggled to detect factual errors which are semantically or pragmatically complex (for example, based on incorrect computation or inference).