Progressive Layout Generationwith Render-Grounded Planning
We propose PlayGround, a progressive layout generation framework that adaptively adjusts the design workflow based on the current rendered canvas. Under our six-criterion VLM evaluation, PlayGround performs favorably against existing methods on both Crello and LICA, including zero-shot evaluation on LICA.
Generating editable graphic designs requires arranging visual and textual elements into a coherent composition while preserving their content and editability. Although recent MLLM-based approaches employ iterative generation and visual refinement to reflect the real-world design workflows, their high-level plans are predetermined prior to execution, limiting their adaptability as the design evolves. To address this limitation, we propose PlayGround, a progressive layout generation framework that dynamically adapts to the current rendered canvas. PlayGround establishes a progressive design loop comprising a Progressive Planner, a Layout Generator, and a Visual Verifier, guided by a Structure-Aware Retriever. Given an initial set of unordered elements, the Structure-Aware Retriever initially infers a coarse layout structure to retrieve structurally similar reference designs. Subsequently, at each turn, the Progressive Planner examines the current rendered canvas alongside the remaining elements, forms the next functional group, and plans its composition. The Layout Generator then predicts editable layout attributes, while the Visual Verifier evaluates rendered candidates and provides feedback when refinement is necessary. To train the Progressive Planner with step-wise supervision, we construct a two-stage data generation pipeline that extracts functional groups and turn-level planning trajectories from completed designs. Experiments on Crello and LICA demonstrate the effectiveness of PlayGround over existing methods. Our code and weights will be publicly released.
From completed designs to functional groups and group-wise turn-level planning supervision. The two-stage pipeline produces group regions, cumulative canvas states, and group-wise turn-level plans, which are used to train the Structure-Aware Retriever, Progressive Planner, and Layout Generator.
An MLLM annotator groups elements that should be planned and positioned together, assigns functional roles, and determines a dependency-aware execution order. Group regions and cumulative canvas states are then derived.
With grouping and execution order fixed, before-and-after renders provide visual evidence for each group’s placement. The annotator produces grouping, execution, and layout rationales together with a qualitative layout instruction.
The Retriever infers a coarse layout structure from the input elements as functional group role–region pairs. It retrieves five references using role-aware spatial similarity, canvas aspect ratio, and text–image content counts. These references remain fixed throughout generation.
The Planner observes the current canvas and remaining elements, forms a functional group, and plans its composition. In the example below, it uses the available upper-left space for text, while the Static Planner, which must produce the entire plan at once and therefore imagine the intermediate canvases, targets a central placard already occupied by photographs.
The Layout Generator predicts element-level attributes for the current group while keeping previously placed elements unchanged. It generates multiple candidate layouts because a group-level plan can admit different rendered outcomes.
Five candidate renders from the pipeline overview.
The Verifier selects a rendered candidate and either accepts it or returns a revised layout instruction for regeneration. Up to three verification rounds are allowed; if none is accepted, it makes a final selection among the per-round finalists.
Table 1. Six VLM-based criteria (mean ± sample standard deviation) and two rule-based metrics. VLM scores range from 1 to 5; higher is better. T.Occ. (text occlusion) is the fraction of text glyph pixels hidden by elements stacked above them; Rea. (text unreadability) measures how visually busy the background beneath the text is. Lower is better for both.
| Method | VLM-based metrics ↑ | Rule-based metrics ↓ | ||||||
|---|---|---|---|---|---|---|---|---|
| (i) Composition |
(ii) Typography |
(iii) Elements |
(iv) Color |
(v) Style |
(vi) Intention |
T.Occ. | Rea. | |
Users may have a design intention without the visual and textual elements needed to realize it. We extend PlayGround to this setting by prepending an element-generation stage.
GPT-5.6 Luna derives the textual elements and visual-element descriptions. Qwen-Image-2.1 generates the corresponding visual elements, which are then composed by PlayGround without modifying the core pipeline.
These examples on Crello demonstrate the modular extensibility of PlayGround as a downstream composition module, from element-to-design to intention-to-design generation.
% Add final publication metadata before publishing.
@misc{playground,
title = {{PlayGround}: Progressive Layout Generation
with Render-Grounded Planning},
author = {Lee, Jaeho and Heo, Junhwan and Park, Junghyun and Moon, Wonjun
and Yang, Eunju and Jang, Seungho and Kim, Seongchan and Kim, Seungryong},
note = {Manuscript}
}
Author names are provided above. Final publication metadata will be added when available; this citation remains a draft.