From Pixels to Policies: Reinforcing Spatial Reasoning in Language Models for Content-Aware Layout Design

TL;DR

LaySPA is a reinforcement learning framework that enhances large language models with explicit spatial reasoning capabilities for content-aware graphic layout design, improving transparency and performance.

Contribution

It introduces a structured textual environment and a multi-objective critique for LLMs, enabling interpretable and effective spatial reasoning in layout design.

Findings

Outperforms larger proprietary LLMs in layout quality

Achieves comparable results to specialized SOTA layout generators

Requires fewer annotated samples and less computation

Abstract

We introduce LaySPA, a reinforcement learning framework that equips large language models (LLMs) with explicit and interpretable spatial reasoning for content-aware graphic layout design. LaySPA addresses two key challenges: LLMs' limited spatial reasoning and the lack of opacity in design decision making. Instead of operating at the pixel level, we reformulate layout design as a policy learning problem over a structured textual spatial environment that explicitly encodes canvas geometry, element attributes, and inter-element relationships. LaySPA produces dual-level outputs comprising interpretable reasoning traces and structured layout specifications, enabling transparent and controllable design decision making. Layout design policy is optimized via a multi-objective spatial critique that decomposes layout quality into geometric validity, relational coherence, and aesthetic consistency, and is trained using relative group optimization to stabilize learning in open-ended design spaces. Experiments demonstrate that LaySPA improves structural validity and visual quality, outperforming larger proprietary LLMs and achieving performance comparable to specialized SOTA layout generators while requiring fewer annotated samples and reduced latency.