Diffusion-based text-to-image generation models have greatly advanced the field of artistic content synthesis. However, current portrait stylization methods typically require fine-tuning models with ...