Controllable Text Generation with Focused Variation
It addresses controllability and diversity issues in text generation for NLP applications, representing an incremental improvement over existing methods.
The paper tackles the dual problems of controllability and diversity in text generation by introducing the Focused-Variation Network (FVN), which uses discrete latent spaces for attributes to achieve state-of-the-art performance on annotated datasets.
This work introduces Focused-Variation Network (FVN), a novel model to control language generation. The main problems in previous controlled language generation models range from the difficulty of generating text according to the given attributes, to the lack of diversity of the generated texts. FVN addresses these issues by learning disjoint discrete latent spaces for each attribute inside codebooks, which allows for both controllability and diversity, while at the same time generating fluent text. We evaluate FVN on two text generation datasets with annotated content and style, and show state-of-the-art performance as assessed by automatic and human evaluations.