Timezone: »

Latent Diffusion for Language Generation
Justin Lovelace · Varsha Kishore · Chao Wan · Eliot Shekhtman · Kilian Weinberger

Wed Dec 13 08:45 AM -- 10:45 AM (PST) @ Great Hall & Hall B1+B2 #528

Diffusion models have achieved great success in modeling continuous data modalities such as images, audio, and video, but have seen limited use in discrete domains such as language. Recent attempts to adapt diffusion to language have presented diffusion as an alternative to existing pretrained language models. We view diffusion and existing language models as complementary. We demonstrate that encoder-decoder language models can be utilized to efficiently learn high-quality language autoencoders. We then demonstrate that continuous diffusion models can be learned in the latent space of the language autoencoder, enabling us to sample continuous latent representations that can be decoded into natural language with the pretrained decoder. We validate the effectiveness of our approach for unconditional, class-conditional, and sequence-to-sequence language generation. We demonstrate across multiple diverse data sets that our latent language diffusion models are significantly more effective than previous diffusion language models. Our code is available at \url{https://github.com/justinlovelace/latent-diffusion-for-language}.

Author Information

Justin Lovelace (Cornell University)
Varsha Kishore (Cornell University)
Chao Wan (Cornell University)
Eliot Shekhtman (Cornell University)
Kilian Weinberger (Cornell University / ASAPP Research)

More from the Same Authors