Controllable Paraphrase Generation with a Syntactic Exemplar
This work addresses controllable text generation for NLP applications, but it is incremental as it builds on prior methods with a new task and dataset.
The paper tackles the problem of generating paraphrases with controlled syntax using a sentential exemplar, and the proposed model achieves improvements over baselines in capturing desirable characteristics.
Prior work on controllable text generation usually assumes that the controlled attribute can take on one of a small set of values known a priori. In this work, we propose a novel task, where the syntax of a generated sentence is controlled rather by a sentential exemplar. To evaluate quantitatively with standard metrics, we create a novel dataset with human annotations. We also develop a variational model with a neural module specifically designed for capturing syntactic knowledge and several multitask training objectives to promote disentangled representation learning. Empirically, the proposed model is observed to achieve improvements over baselines and learn to capture desirable characteristics.