“Zero-Shot Translation Using Diffusion Models”, 2021-11-02 (; similar):
In this work, we show a novel method for neural machine translation (NMT), using a denoising diffusion probabilistic model (DDPM), adjusted for textual data, following recent advances in the field.
We show that it’s possible to translate sentences non-autoregressively using a diffusion model conditioned on the source sentence.
We also show that our model is able to translate between pairs of languages unseen during training (zero-shot learning).