# Adversarial Text Generation Without Reinforcement Learning

arXiv: https://arxiv.org/abs/1810.06640 · HTML: https://mlsanity.com/paper/1810.06640/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Oct 2018.

## Built on by (most cited first, 7 of 7)

- [A survey on text generation using generative adversarial networks](https://mlsanity.com/paper/2212.11119/) (2022)
- [OodGAN: Generative Adversarial Network for Out-of-Domain Data Generation](https://mlsanity.com/paper/2104.02484/) (2021)
- [Regularizing Transformers With Deep Probabilistic Layers](https://mlsanity.com/paper/2108.10764/) (2021)
- [Claim Verification using a Multi-GAN based Model](https://mlsanity.com/paper/2103.08001/) (2021)
- [OptAGAN: Entropy-based finetuning on text VAE-GAN](https://mlsanity.com/paper/2109.00239/) (2021)
- [Forging the Forger: An Attempt to Improve Authorship Verification via Data Augmentation](https://mlsanity.com/paper/2403.11265/) (2024)
- [VicSim: Enhancing Victim Simulation with Emotional and Linguistic Fidelity](https://mlsanity.com/paper/2501.03139/) (2025)

## Abstract (arXiv)

Generative Adversarial Networks (GANs) have experienced a recent surge in popularity, performing competitively in a variety of tasks, especially in computer vision. However, GAN training has shown limited success in natural language processing. This is largely because sequences of text are discrete, and thus gradients cannot propagate from the discriminator to the generator. Recent solutions use reinforcement learning to propagate approximate gradients to the generator, but this is inefficient to train. We propose to utilize an autoencoder to learn a low-dimensional representation of sentences. A GAN is then trained to generate its own vectors in this space, which decode to realistic utterances. We report both random and interpolated samples from the generator. Visualization of sentence vectors indicate our model correctly learns the latent space of the autoencoder. Both human ratings and BLEU scores show that our model generates realistic text against competitive baselines.