Pengyu Zeng

Reinforcement Learning (1)

floor plan generation (2) reinforcement learning (2) architectural design (2) multimodal learning (1) knowledge base (1) reinforcement learning from human feedback (1) domain knowledge (1) diffusion model (1) reward model (1) adversarial example (1) retrieval-augmented generation (1) 3d rendering (1) 3d point cloud (1) group relative policy optimization (1) preference score (1) single-view 3d reconstruction (1) query formulation (1) building generation (1) window-to-wall ratio (1) single-view generation (1)