Agentic Designer: Progressive Multi-Agent Collaboration for Structure-Aware Interior Layout Generation
Quick Answer
Agentic Designer introduces a multi-agent framework for generating interior layouts that adhere to architectural constraints, significantly outperforming existing methods.
Quick Take
By utilizing a Progressive Consensus Mechanism, it ensures geometric validation at each stage, resulting in improved structural integrity and functional coherence. The framework is benchmarked against over 18,000 samples, demonstrating superior performance in layout generation.
Key Points
- Agentic Designer employs a Generator, Evaluator, and Refiner for layout generation.
- The framework prevents error accumulation through iterative geometric validation.
- InStruct benchmark includes over 18,000 high-quality, parametrically annotated samples.
- Significant improvements in structural adherence and design coherence were observed.
- User studies confirm Agentic Designer's superiority over state-of-the-art methods.
DeepSignal Analysis
What happened
Agentic Designer is a new multi-agent framework designed for generating interior layouts that meet architectural constraints. It employs a Progressive Consensus Mechanism for iterative geometric validation, which helps prevent structural errors during layout generation.
Key evidence
- The framework coordinates three specialized agents: a Generator, an Evaluator, and a Refiner, to ensure that each layout proposal is verified and adjusted iteratively.
- Agentic Designer was benchmarked against over 18,000 samples, demonstrating superior performance in generating layouts that adhere to structural and functional requirements.
- The proposed framework addresses limitations of existing methods that often result in structural collisions and infeasible arrangements due to a lack of intermediate geometric constraint verification.
Why it matters
This advancement in interior layout generation is significant as it addresses longstanding challenges in automated spatial design. By ensuring that layouts adhere to architectural constraints, the framework could enhance the efficiency and reliability of design processes in architecture and interior design, potentially impacting various industries reliant on automated design tools.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Generating realistic interior furniture layouts that strictly adhere to architectural constraints (e.g., walls, doors, and windows) remains a fundamental challenge in automated spatial design. Existing approaches, primarily based on one-shot generation using diffusion models or Large Language Models (LLMs), lack explicit mechanisms for intermediate geometric constraint verification, often resulting in structural collisions and functionally infeasible arrangements under complex room constraints. To address these challenges, we propose Agentic Designer, a progressive, multi-agent framework that formulates structure-aware interior layout generation as an iterative and constraint-verified decision process. By decomposing layout synthesis into modular stages of proposal, verification, and adjustment, the framework coordinates three specialized agents, a Generator, an Evaluator, and a Refiner, through a Progressive Consensus Mechanism. This mechanism enforces stepwise geometric validation and correction before each placement is committed, thereby preventing error accumulation. To facilitate this structure-aware paradigm and standardize evaluation, we establish InStruct, a comprehensive benchmark that integrates a dataset comprising over 18,000 high-quality, parametrically annotated samples with a novel suite of structure-centric metrics. Extensive quantitative evaluations, qualitative analyses, and user studies show that Agentic Designer significantly outperforms state-of-the-art methods, demonstrating substantial improvements in strict structural adherence and functional design coherence.
| Comments: | TPAMI 2026 |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.20866 [cs.CV] |
| (or arXiv:2607.20866v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.20866 arXiv-issued DOI via DataCite (pending registration) |
|
| Related DOI: | https://doi.org/10.1109/TPAMI.2026.3711762
DOI(s) linking to related resources |
Submission history
From: Zhihua Xu [view email]
[v1]
Thu, 23 Jul 2026 02:46:11 UTC (7,128 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.