4

I am wondering why there has not been more usage of GANs for NLP. I know there has been research on the subject (The Google Scholar page for the subject is here).

Are there any specific reasons why GANs do not work for NLP specifically VQGAN + CLIP variants? I do not understand why most text generated by AI is done through predicting the next letter or word in a sequence with RNNs when GANs have had so much success generating deep fakes and the such instead of say, predicting the next pixel.

catasaurus
  • 145
  • 1
  • 7

1 Answers1

6

A few reasons:

  1. Transformers are amazing at text generation already (e.g. GPT-3 which almost passes the Turing-Test)
  2. The original GAN requires a continuous data representation (e.g. images) instead of a discrete one (e.g. text), so that slight error signals can be used for learning.
  3. Empirically speaking, GANs don't seem to work that well on non-image data. I recently applied them to regular tabular data but found auto-encoders much more useful.

P.S. Also it's my intuition that GAN's and Denoising Diffusion Models are popular in the image generation community because they both offer mechanisms that can prevent "blurriness" in the generated images (either the adversary or extra denoising iterations).

profPlum
  • 496
  • 2
  • 10