ReDesign: Recovering Editable Design Structures from Images via Agentic Decomposition

Hugging Face Daily Papers Papers
agentic-framework design-reconstruction editability typography vector-geometry layer-hierarchy figma-benchmark

Summary

ReDesign is an agentic framework that recovers editable layer hierarchies from raster images by selecting and composing specialized tools across modalities, introducing graceful verification to prevent error accumulation. It also introduces the FigmaEditReplay Benchmark for evaluating editability at scale, achieving high visual fidelity and superior editability over baselines.

Recovering an editable design file from a raster image is a common and costly bottleneck in modern design workflows, yet remains challenging since editability depends on recovering multi-modal attributes, such as typography, vector geometry, colors, grouping, and layer ordering. We present ReDesign, an agentic framework that grows an editable layer hierarchy by selecting and composing specialized tools across modalities. To keep this long decision process reliable despite imperfect tool outputs, we introduce graceful verification at each expansion, which provides local accept, prune, or retry feedback that prevents error accumulation and avoids large scale reruns. To evaluate editability at scale, we introduce the Figma Edit Replay Benchmark, consisting of 909 raw Figma files and 14,796 controlled edit instructions that replay edits on reconstructed outputs. Across this benchmark and standard reconstruction metrics, ReDesign achieves strong visual fidelity while delivering the highest editability across layout, color, and text edits, outperforming layered decomposition baselines and serial tool use pipelines.
Original Article
View Cached Full Text

Cached at: 07/29/26, 07:51 AM

Paper page - ReDesign: Recovering Editable Design Structures from Images via Agentic Decomposition

Source: https://huggingface.co/papers/2607.25565

Abstract

Recoveringaneditabledesignfilefromarasterimageisacommonandcostlybottleneckinmoderndesignworkflows,yetremainschallengingsinceeditabilitydependsonrecoveringmulti-modalattributes,suchastypography,vectorgeometry,colors,grouping,andlayerordering.WepresentReDesign,anagenticframeworkthatgrowsaneditablelayerhierarchybyselectingandcomposingspecializedtoolsacrossmodalities.Tokeepthislongdecisionprocessreliabledespiteimperfecttooloutputs,weintroducegracefulverificationateachexpansion,whichprovideslocalaccept,prune,orretryfeedbackthatpreventserroraccumulationandavoidslargescalereruns.Toevaluateeditabilityatscale,weintroducetheFigmaEditReplayBenchmark,consistingof909rawFigmafilesand14,796controllededitinstructionsthatreplayeditsonreconstructedoutputs.Acrossthisbenchmarkandstandardreconstructionmetrics,ReDesignachievesstrongvisualfidelitywhiledeliveringthehighesteditabilityacrosslayout,color,andtextedits,outperforminglayereddecompositionbaselinesandserialtoolusepipelines.

View arXiv pageView PDFProject pageGitHub0Add to collection

Get this paper in your agent:

hf papers read 2607\.25565

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.25565 in a model README.md to link it from this page.

Datasets citing this paper1

#### Jintae-Park/ReDesign-Figma909 Updatedabout 4 hours ago • 130 • 1

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.25565 in a Space README.md to link it from this page.

Collections including this paper1

Similar Articles

Editable Visual Design

Hugging Face Daily Papers

A coding agent guided by vision-language models generates editable layered designs by synthesizing visual assets and refining HTML/CSS layouts, enabling post-editing with precise control.

Is This Edit Correct? A Multi-Dimensional Benchmark for Reasoning-Aware Image Editing

Hugging Face Daily Papers

This paper introduces RE-Edit, a benchmark for evaluating image editing systems across five reasoning dimensions (physical, environmental, cultural, causal, referential) to assess logical consistency beyond visual plausibility. The benchmark includes 1,000 samples and evaluates ten open-source and two commercial models, showing that even advanced systems struggle with implicit multi-dimensional reasoning.