Generative Floor Plan Design with LLMs via Reinforcement Learning with Verifiable Rewards
Quick Answer
This paper shows that A novel approach for floor plan generation utilizes a fine-tuned large language model (LLM) and reinforcement learning with verifiable rewards (RLVR) to adhere to topological and numerical constraints, achieving over 94% improvement in compatibility compared to existing methods.
Quick Take
This method enhances the realism, compatibility, and diversity of generated plans, demonstrating the potential of in constraint-driven design tasks.
Key Points
- Introduces a text-based floor plan generation approach using LLMs and RLVR.
- Achieves at least a 94% relative reduction in compatibility issues.
- Generates plans that meet user-defined connectivity and numerical constraints.
- Outperforms existing methods in realism, compatibility, and diversity metrics.
- Demonstrates broader applications for text-based generative modeling.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2605. 14117v1 Announce Type: new Abstract: An AI system for professional floor plan design must precisely control room dimensions and areas while respecting the desired connectivity between rooms and maintaining functional and aesthetic quality. Existing generative approaches focus primarily on respecting the requested connectivity between rooms, but do not support generating floor plans that respect numerical constraints.
We introduce a text-based floor plan generation approach that fine-tunes a (LLM) on real plans and then applies reinforcement learning with verifiable rewards (RLVR) to improve adherence to topological and numerical constraints while discouraging invalid or overlapping outputs. Furthermore, we design a set of constraint adherence metrics to systematically measure how generated floor plans align with user-defined constraints.
Our model generates floor plans that satisfy user-defined connectivity and numerical constraints and outperforms existing methods on Realism, Compatibility, and Diversity metrics. Across all tasks, our approach achieves at least a 94% relative reduction in Compatibility compared with existing methods. Our results demonstrate that LLMs can effectively handle constraints in this setting, suggesting broader applications for text-based generative modeling.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.