Generating (Factual?) Narrative Summaries of RCTs: Experiments with Neural Multi-Document Summarization
This addresses the problem of automating evidence synthesis for medical professionals, but it is incremental as it builds on existing summarization methods with minor enhancements.
The paper tackled generating narrative biomedical evidence summaries from multiple trial reports using neural models, finding that while summaries were fluent and relevant, they often lacked factual accuracy, with proposed domain-specific strategies only modestly improving accuracy.
We consider the problem of automatically generating a narrative biomedical evidence summary from multiple trial reports. We evaluate modern neural models for abstractive summarization of relevant article abstracts from systematic reviews previously conducted by members of the Cochrane collaboration, using the authors conclusions section of the review abstract as our target. We enlist medical professionals to evaluate generated summaries, and we find that modern summarization systems yield consistently fluent and relevant synopses, but that they are not always factual. We propose new approaches that capitalize on domain-specific models to inform summarization, e.g., by explicitly demarcating snippets of inputs that convey key findings, and emphasizing the reports of large and high-quality trials. We find that these strategies modestly improve the factual accuracy of generated summaries. Finally, we propose a new method for automatically evaluating the factuality of generated narrative evidence syntheses using models that infer the directionality of reported findings.