The Free Transformer
This work addresses the need for enhanced generative modeling in AI, though it appears incremental as it builds on existing Transformer architectures.
The authors tackled the problem of improving generative processes in decoder Transformers by conditioning on unsupervised random latent variables, resulting in substantial improvements on downstream tasks.
We propose an extension of the decoder Transformer that conditions its generative process on random latent variables which are learned without supervision thanks to a variational procedure. Experimental evaluations show that allowing such a conditioning translates into substantial improvements on downstream tasks.