# Incorporating Reinforced Adversarial Learning in Autoregressive Image Generation

arXiv: https://arxiv.org/abs/2007.09923 · HTML: https://mlsanity.com/paper/2007.09923/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jul 2020.

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

- [Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunities](https://mlsanity.com/paper/2505.02567/) (2025)
- [Autoregressive Models in Vision: A Survey](https://mlsanity.com/paper/2411.05902/) (2024)
- [The Image Local Autoregressive Transformer](https://mlsanity.com/paper/2106.02514/) (2021)
- [From Prediction to Perfection: Introducing Refinement to Autoregressive Image Generation](https://mlsanity.com/paper/2505.16324/) (2025)
- [FashionSearchNet-v2: Learning Attribute Representations with Localization for Image Retrieval with Attribute Manipulation](https://mlsanity.com/paper/2111.14145/) (2021)
- [Uni-ViGU: Towards Unified Video Generation and Understanding via A Diffusion-Based Video Generator](https://mlsanity.com/paper/2604.08121/) (2026)
- [RAVEN: Rethinking Adversarial Video Generation with Efficient Tri-plane Networks](https://mlsanity.com/paper/2401.06035/) (2024)

## Abstract (arXiv)

Autoregressive models recently achieved comparable results versus state-of-the-art Generative Adversarial Networks (GANs) with the help of Vector Quantized Variational AutoEncoders (VQ-VAE). However, autoregressive models have several limitations such as exposure bias and their training objective does not guarantee visual fidelity. To address these limitations, we propose to use Reinforced Adversarial Learning (RAL) based on policy gradient optimization for autoregressive models. By applying RAL, we enable a similar process for training and testing to address the exposure bias issue. In addition, visual fidelity has been further optimized with adversarial loss inspired by their strong counterparts: GANs. Due to the slow sampling speed of autoregressive models, we propose to use partial generation for faster training. RAL also empowers the collaboration between different modules of the VQ-VAE framework. To our best knowledge, the proposed method is first to enable adversarial learning in autoregressive models for image generation. Experiments on synthetic and real-world datasets show improvements over the MLE trained models. The proposed method improves both negative log-likelihood (NLL) and Fr'echet Inception Distance (FID), which indicates improvements in terms of visual quality and diversity. The proposed method achieves state-of-the-art results on Celeba for 64 × 64 image resolution, showing promise for large scale image generation.