Research map / Multi-agent systems and games / LLM agents
Agentic Designer: Progressive Multi-Agent Collaboration for Structure-Aware Interior Layout Generation
Zhijing Yang, H H Lin, Zhihua Xu, Hui Li, Keze Wang, L Y Liang, Tianshui Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Evaluator and Refiner agents to the Generator (ablation) | Agentic Designer · InStruct Bedroom | SVR (%): 21.14 -> 7.14 | Improved |
| add Evaluator and Refiner agents to the Generator (ablation) | Agentic Designer · InStruct Bedroom | CS: 5.31 -> 3.10 | Improved |
| add Evaluator and Refiner agents to the Generator (ablation) | Agentic Designer · InStruct Living Room | SVR (%): 20.00 -> 16.85 | Improved |
| add Evaluator and Refiner agents (user study) (ablation) | Agentic Designer · User study ablation | CR/LR/PR: CR 3.63 -> 4.44, LR 3.15 -> 4.39, PR 3.70 -> 4.49 | Improved |
| use progressive step-wise verification instead of global refinement (ablation) | Agentic Designer · InStruct Bedroom | SVR (%): 17.74 -> 7.14 | Improved |
| use progressive step-wise verification instead of global refinement (ablation) | Agentic Designer · InStruct Bedroom | CS: 3.99 -> 3.10 | Improved |
| use progressive step-wise verification instead of global refinement (ablation) | Agentic Designer · InStruct Living Room | CS: 2.54 -> 3.27 | Hurt |
| condition Generator on verified history instead of unverified (ablation) | Generator · InStruct Living Room | SVR/BVC/CS: 20.00%/0.38/3.42 -> 17.97%/0.33/2.19 | Improved |
| condition Generator on verified history instead of unverified (ablation) | Generator · InStruct Bedroom | SVR/BVC/CS: 21.14%/0.29/5.31 -> 17.38%/0.25/3.14 | Improved |
| raise max refinement iterations Nmax from 1 to 2 | Agentic Designer · InStruct Living Room | SVR (%): 16.85 -> 10.04 | Improved |
| raise max refinement iterations Nmax from 2 to 4 | Agentic Designer · InStruct Living Room | SVR (%): 10.04 -> 8.43 | No clear effect |
| raise max refinement iterations Nmax from 1 to 4 | Agentic Designer · InStruct Bedroom | SVR (%): 7.14 -> 4.74 | Improved |
Closest papers by meaning
- Constrained Layout Generation with Factor Graphs (CVPR 2024 · similarity 0.88)
- OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization (NeurIPS 2025 · similarity 0.87)
- MaskPLAN: Masked Generative Layout Planning from Partial Input (CVPR 2024 · similarity 0.87)
- Tokenization Allows Multimodal Large Language Models to Understand, Generate and Edit Architectural Floor Plans (CVPR 2026 · similarity 0.86)
- Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning (NeurIPS 2025 · similarity 0.86)
- Towards Aligned Layout Generation via Diffusion Model with Aesthetic Constraints (ICLR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
