The field of generative models is rapidly advancing, with a focus on creating realistic 3D environments and personalized content. Recent developments have led to the creation of novel methods for generating indoor scenes, 3D building models, and underground environments. These models are capable of capturing fine-grained object placements, ensuring structurally coherent and physically plausible scene generation. The use of hierarchical approaches, diffusion models, and large language models has improved the efficiency and scalability of these methods, making them suitable for large-scale deployment. Notable papers in this area include: DecoMind, which introduces a system for generating interior design layouts based on user inputs. HLG, which proposes a novel method for fine-grained 3D scene generation using a coarse-to-fine hierarchical approach. HLLM-Creator, which presents a hierarchical LLM framework for efficient user interest modeling and personalized content generation. SAT-SKYLINES, which generates 3D building models from satellite imagery and coarse geometric priors. SemLayoutDiff, which synthesizes diverse 3D indoor scenes using a categorical diffusion model. PLUME, which creates 3D underground environments using a procedural generation framework.