# Disentangled Inference for GANs with Latently Invertible Autoencoder

arXiv: https://arxiv.org/abs/1906.08090 · HTML: https://mlsanity.com/paper/1906.08090/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Jun 2019.

## Built on by (most cited first, 10 of 10)

- [GAN Inversion: A Survey](https://mlsanity.com/paper/2101.05278/) (2021)
- [Normalized Avatar Synthesis Using StyleGAN and Perceptual Refinement](https://mlsanity.com/paper/2106.11423/) (2021)
- [Harnessing Diffusion-Yielded Score Priors for Image Restoration](https://mlsanity.com/paper/2507.20590/) (2025)
- [Face Generation and Editing with StyleGAN: A Survey](https://mlsanity.com/paper/2212.09102/) (2022)
- [High-fidelity GAN Inversion with Padding Space](https://mlsanity.com/paper/2203.11105/) (2022)
- [In-Domain GAN Inversion for Faithful Reconstruction and Editability](https://mlsanity.com/paper/2309.13956/) (2023)
- [CCR: Facial Image Editing with Continuity, Consistency and Reversibility](https://mlsanity.com/paper/2209.10734/) (2022)
- [IGAN: Inferent and Generative Adversarial Networks](https://mlsanity.com/paper/2109.13360/) (2021)
- [Conditioning Generative Latent Optimization for Sparse-View CT Image Reconstruction](https://mlsanity.com/paper/2307.16670/) (2023)
- [Variational Rectification Inference for Learning with Noisy Labels](https://mlsanity.com/paper/2603.17255/) (2026)

## Abstract (arXiv)

Generative Adversarial Networks (GANs) play an increasingly important role in machine learning. However, there is one fundamental issue hindering their practical applications: the absence of capability for encoding real-world samples. The conventional way of addressing this issue is to learn an encoder for GAN via Variational Auto-Encoder (VAE). In this paper, we show that the entanglement of the latent space for the VAE/GAN framework poses the main challenge for encoder learning. To address the entanglement issue and enable inference in GAN we propose a novel algorithm named Latently Invertible Autoencoder (LIA). The framework of LIA is that an invertible network and its inverse mapping are symmetrically embedded in the latent space of VAE. The decoder of LIA is first trained as a standard GAN with the invertible network and then the partial encoder is learned from a disentangled autoencoder by detaching the invertible network from LIA, thus avoiding the entanglement problem caused by the random latent space. Experiments conducted on the FFHQ face dataset and three LSUN datasets validate the effectiveness of LIA/GAN.