R$^3$L: Reasoning 3D Layouts from Relative Spatial Relations

TL;DR

R3L introduces a framework that enhances the reliability of 3D layout generation by addressing errors in multi-hop relative spatial reasoning through invariant decomposition, self-consistency, and spatial optimization.

Contribution

It proposes novel methods for invariant spatial decomposition and consistent spatial imagination to improve multi-hop reasoning in 3D layout generation.

Findings

Produces more physically feasible layouts

Achieves higher semantic consistency

Addresses frame-induced reasoning errors

Abstract

Relative spatial relations provide a compact representation of spatial structure and are fundamental to relative spatial reasoning in 3D layout generation. Recent works leverage Multimodal Large Language Models (MLLMs) to infer such relations, but the inferred relations are often unreliable and are typically handled with post-hoc heuristics. In this paper, we propose R$^3$L, a general framework that improves the reliability and consistency of relative spatial reasoning for 3D layout generation. Our key motivation is that multi-hop reasoning requires repeated reference-frame transformations, which accumulate errors in inferred relations and lead to semantic and metric drift. To mitigate this, we propose invariant spatial decomposition to break coupled relation chains, and consistent spatial imagination to promote self-consistency through an imagine-and-revise loop. We further introduce supportive spatial optimization to ease pose optimization via global-to-local coordinate re-parameterization. Extensive experiments across diverse scene types and instructions demonstrate that R$^3$L produces more physically feasible and semantically consistent layouts. Notably, our analysis shows that resolving frame-induced inconsistencies is crucial for reliable multi-hop relative spatial reasoning. The code is available at https://github.com/Neal2020GitHub/R3L.