ColdGANs: Taming Language GANs with Cautious Sampling Strategies

TL;DR

ColdGANs introduce cautious sampling strategies to stabilize language GAN training, outperforming traditional MLE methods and state-of-the-art models across multiple text generation tasks.

Contribution

This paper presents ColdGANs, a novel approach that uses mode-focused sampling to improve the stability and performance of GAN-based language generation.

Findings

ColdGANs outperform MLE in several text generation tasks.

Cautious sampling leads to more stable GAN training.

Significant improvements over state-of-the-art methods on three tasks.

Abstract

Training regimes based on Maximum Likelihood Estimation (MLE) suffer from known limitations, often leading to poorly generated text sequences. At the root of these limitations is the mismatch between training and inference, i.e. the so-called exposure bias, exacerbated by considering only the reference texts as correct, while in practice several alternative formulations could be as good. Generative Adversarial Networks (GANs) can mitigate those limitations but the discrete nature of text has hindered their application to language generation: the approaches proposed so far, based on Reinforcement Learning, have been shown to underperform MLE. Departing from previous works, we analyze the exploration step in GANs applied to text generation, and show how classical sampling results in unstable training. We propose to consider alternative exploration strategies in a GAN framework that we name ColdGANs, where we force the sampling to be close to the distribution modes to get smoother learning dynamics. For the first time, to the best of our knowledge, the proposed language GANs compare favorably to MLE, and obtain improvements over the state-of-the-art on three generative tasks, namely unconditional text generation, question generation, and abstractive summarization.