MaskGAN: Better Text Generation via Filling in the______

TL;DR

This paper introduces MaskGAN, a novel actor-critic conditional GAN approach for text generation that improves sample quality by filling in missing text conditioned on context, surpassing traditional likelihood-based models.

Contribution

The paper presents MaskGAN, the first application of GANs to fill-in-the-blank text tasks, enhancing sample realism over standard maximum likelihood models.

Findings

MaskGAN produces more realistic text samples.

GAN-based training improves sample quality.

Perplexity is not a sole indicator of text quality.

Abstract

Neural text generation models are often autoregressive language models or seq2seq models. These models generate text by sampling words sequentially, with each word conditioned on the previous word, and are state-of-the-art for several machine translation and summarization benchmarks. These benchmarks are often defined by validation perplexity even though this is not a direct measure of the quality of the generated text. Additionally, these models are typically trained via maxi- mum likelihood and teacher forcing. These methods are well-suited to optimizing perplexity but can result in poor sample quality since generating text requires conditioning on sequences of words that may have never been observed at training time. We propose to improve sample quality using Generative Adversarial Networks (GANs), which explicitly train the generator to produce high quality samples and have shown a lot of success in image generation. GANs were originally designed to output differentiable values, so discrete language generation is challenging for them. We claim that validation perplexity alone is not indicative of the quality of text generated by a model. We introduce an actor-critic conditional GAN that fills in missing text conditioned on the surrounding context. We show qualitatively and quantitatively, evidence that this produces more realistic conditional and unconditional text samples compared to a maximum likelihood trained model.