Diffusion Models in NLP: A Survey

Abstract: Diffusion models have become a powerful family of deep generative models, with record-breaking performance in many applications. This paper first gives an overview and derivation of the basic theory of diffusion models, then reviews the research results of diffusion models in the field of natural language processing, from text generation, text-driven image generation and other four aspects, and analyzes and summarizes the relevant literature materials sorted out, and finally records the experience and feelings of this topic literature review research.