# Self-Adversarial Learning with Comparative Discrimination for Text Generation

arXiv: https://arxiv.org/abs/2001.11691 · HTML: https://mlsanity.com/paper/2001.11691/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jan 2020.

## Built on by (most cited first, 11 of 11)

- [Inverse-RLignment: Large Language Model Alignment from Demonstrations through Inverse Reinforcement Learning](https://mlsanity.com/paper/2405.15624/) (2024)
- [Latent Imitator: Generating Natural Individual Discriminatory Instances for Black-Box Fairness Testing](https://mlsanity.com/paper/2305.11602/) (2023)
- [A survey on text generation using generative adversarial networks](https://mlsanity.com/paper/2212.11119/) (2022)
- [To Beam Or Not To Beam: That is a Question of Cooperation for Language GANs](https://mlsanity.com/paper/2106.06363/) (2021)
- [Learning from Perturbations: Diverse and Informative Dialogue Generation with Inverse Adversarial Training](https://mlsanity.com/paper/2105.15171/) (2021)
- [Reinforcement Learning for Generative AI: A Survey](https://mlsanity.com/paper/2308.14328/) (2023)
- [SGBANet: Semantic GAN and Balanced Attention Network for Arbitrarily Oriented Scene Text Recognition](https://mlsanity.com/paper/2207.10256/) (2022)
- [Generative Cooperative Networks for Natural Language Generation](https://mlsanity.com/paper/2201.12320/) (2022)
- [Improved Training of Mixture-of-Experts Language GANs](https://mlsanity.com/paper/2302.11875/) (2023)
- [Adaptive Bridge between Training and Inference for Dialogue](https://mlsanity.com/paper/2110.11560/) (2021)
- [ComGAN: Toward GANs Exploiting Multiple Samples](https://mlsanity.com/paper/2304.12098/) (2023)

## Abstract (arXiv)

Conventional Generative Adversarial Networks (GANs) for text generation tend to have issues of reward sparsity and mode collapse that affect the quality and diversity of generated samples. To address the issues, we propose a novel self-adversarial learning (SAL) paradigm for improving GANs' performance in text generation. In contrast to standard GANs that use a binary classifier as its discriminator to predict whether a sample is real or generated, SAL employs a comparative discriminator which is a pairwise classifier for comparing the text quality between a pair of samples. During training, SAL rewards the generator when its currently generated sentence is found to be better than its previously generated samples. This self-improvement reward mechanism allows the model to receive credits more easily and avoid collapsing towards the limited number of real samples, which not only helps alleviate the reward sparsity issue but also reduces the risk of mode collapse. Experiments on text generation benchmark datasets show that our proposed approach substantially improves both the quality and the diversity, and yields more stable performance compared to the previous GANs for text generation.