Diversity Enhanced Table-to-Text Generation via Logic-Type Control

Anonymous

Abstract: Generating natural language statements to convey logical inferences from tabular data (i.e., Logical NLG) is a process with one input and a variety of valid outputs. This characteristic underscores the ability for a method to produce a diverse set of valid outputs.We propose a simple yet effective diversity enhancing scheme that builds upon an inherent property of the statements, their logic-types, by using a type-controlled table-to-text generation model. We demonstrate, through extensive automatic and human evaluations over the two publicly available Logical NLG datasets, that our method is able to surpass the strongest baselines along the quality-diversity plane, all while allowing users to effectively control the type of the generated statement.

Paper Type: short

Research Area: Generation