Token Manipulation Generative Adversarial Network for Text Generation
This addresses a specific limitation in conditional text generation for NLP applications, but appears incremental as it builds on existing MaskGAN and RL methods.
The paper tackles the limitation of having to specify blanks in conditional text generation by decomposing the problem into make-a-blank and fill-in-the-blank tasks, extending token manipulations and using hierarchical multi-agent RL with adversarial learning, resulting in good performance without compromising quality and diversity.
MaskGAN opens the query for the conditional language model by filling in the blanks between the given tokens. In this paper, we focus on addressing the limitations caused by having to specify blanks to be filled. We decompose conditional text generation problem into two tasks, make-a-blank and fill-in-the-blank, and extend the former to handle more complex manipulations on the given tokens. We cast these tasks as a hierarchical multi agent RL problem and introduce a conditional adversarial learning that allows the agents to reach a goal, producing realistic texts, in cooperative setting. We show that the proposed model not only addresses the limitations but also provides good results without compromising the performance in terms of quality and diversity.