Generative Floor Plan Design with LLMs via Reinforcement Learning with Verifiable Rewards

TL;DR

This paper presents a novel method using fine-tuned large language models and reinforcement learning with verifiable rewards to generate floor plans that meet both topological and numerical constraints.

Contribution

It introduces a text-based floor plan generation approach that enforces constraints and improves adherence through reinforcement learning with verifiable rewards.

Findings

Generated floor plans satisfy user-defined constraints more effectively.

Our method outperforms existing approaches on Realism, Compatibility, and Diversity metrics.

Achieves at least 94% reduction in Compatibility issues compared to prior methods.

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 large language model (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.