3D-Layout-R1: Structured Reasoning for Language-Instructed Spatial Editing

TL;DR

This paper presents a structured reasoning framework for language-guided spatial editing that enhances interpretability and accuracy in scene layout modifications by reasoning over scene graphs.

Contribution

It introduces a novel scene-graph reasoning approach for text-conditioned spatial editing, improving spatial accuracy and interpretability over existing methods.

Findings

15% average improvement in IoU

25% reduction in center-distance error

Up to 20% higher mIoU compared to SOTA zero-shot LLMs

Abstract

Large Language Models (LLMs) and Vision Language Models (VLMs) have shown impressive reasoning abilities, yet they struggle with spatial understanding and layout consistency when performing fine-grained visual editing. We introduce a Structured Reasoning framework that performs text-conditioned spatial layout editing via scene-graph reasoning. Given an input scene graph and a natural-language instruction, the model reasons over the graph to generate an updated scene graph that satisfies the text condition while maintaining spatial coherence. By explicitly guiding the reasoning process through structured relational representations, our approach improves both interpretability and control over spatial relationships. We evaluate our method on a new text-guided layout editing benchmark encompassing sorting, spatial alignment, and room-editing tasks. Our training paradigm yields an average 15% improvement in IoU and 25% reduction in center-distance error compared to Chain of Thought Fine-tuning (CoT-SFT) and vanilla GRPO baselines. Compared to SOTA zero-shot LLMs, our best models achieve up to 20% higher mIoU, demonstrating markedly improved spatial precision.