Zero-Shot Translation using Diffusion Models
arXiv:2111.01471
Abstract
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).
arXiv commentpreprint
References in corpus (7)
- Adam: A Method for Stochastic Optimization
- Denoising Diffusion Probabilistic Models
- Diffusion Models Beat GANs on Image Synthesis
- NVAE: A Deep Hierarchical Variational Autoencoder
- Generating Diverse High-Fidelity Images with VQ-VAE-2
- Structured Denoising Diffusion Models in Discrete State-Spaces
- Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions