Log In Sign Up

Can the Transformer Be Used as a Drop-in Replacement for RNNs in Text-Generating GANs?

by   Kevin Blin, et al.

In this paper we address the problem of fine-tuned text generation with a limited computational budget. For that, we use a well-performing text generative adversarial network (GAN) architecture - Diversity-Promoting GAN (DPGAN), and attempted a drop-in replacement of the LSTM layer with a self-attention-based Transformer layer in order to leverage their efficiency. The resulting Self-Attention DPGAN (SADPGAN) was evaluated for performance, quality and diversity of generated text and stability. Computational experiments suggested that a transformer architecture is unable to drop-in replace the LSTM layer, under-performing during the pre-training phase and undergoing a complete mode collapse during the GAN tuning phase. Our results suggest that the transformer architecture need to be adapted before it can be used as a replacement for RNNs in text-generating GANs.


page 1

page 2

page 3

page 4


Keyword Transformer: A Self-Attention Model for Keyword Spotting

The Transformer architecture has been successful across many domains, in...

SALSA-TEXT : self attentive latent space based adversarial text generation

Inspired by the success of self attention mechanism and Transformer arch...

Improved Transformer for High-Resolution GANs

Attention-based models, exemplified by the Transformer, can effectively ...

Empirical Analysis of Overfitting and Mode Drop in GAN Training

We examine two key questions in GAN training, namely overfitting and mod...

ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing

We address the problem of finding realistic geometric corrections to a f...

Text Generation with Deep Variational GAN

Generating realistic sequences is a central task in many machine learnin...

Evade the Trap of Mediocrity: Promoting Diversity and Novelty in Text Generation via Concentrating Attention

Recently, powerful Transformer architectures have proven superior in gen...