Guidance Is Not a Hyperparameter: Learning Dynamic Control in Diffusion Language Models

ArXi:2605.07701v1 Announce Type: new Classifier-Free Guidance (CFG) is a widely used mechanism for controlling diffusion-based generative models, yet its guidance scale is typically treated as a fixed hyperparameter throughout generation. This static design yields a suboptimal controllability and quality tradeoff, as the optimal degree of guidance varies across tasks and across different stages of the diffusion process, especially in NLP domain.