InfoEdit: Probing Global Layout Reasoning in Infographic Editing

2026-09-27

AI Summary

This study addresses the challenge that multimodal models struggle with global layout reasoning, or reflow, during infographic editing. To investigate this, the work formally defines and quantifies the task for the first time, constructing a benchmark dataset comprising one thousand infographics. It further introduces a reflow-aware evaluation protocol alongside a code-level editing comparison framework, systematically assessing approaches that integrate pixel-level and code-level image editing with structured logic modeling. The findings reveal significant limitations in existing models on this task, with most achieving success rates below 7%, while only GPT-Image-2 surpasses 60%. Ultimately, this research establishes a critical benchmark and opens new directions for advancing structured visual content editing.

Abstract

Multimodal foundation models edit natural photographs at production quality, yet the same models struggle with structured visual content such as infographics. Unlike photographs, infographics encode information through logical relations; editing one element often requires surrounding elements to be adapted. We refer to this global layout reasoning capability as reflow. Existing image-editing benchmarks neither provide a dedicated setting for structured visual content nor evaluate the reflow capability. We introduce InfoEdit, a novel benchmark of 1,000 infographics across eight logical-relation families, paired with 4,000 editing instructions across four editing tasks, and a reflow-aware evaluation protocol. Across eight frontier editors, only GPT-Image-2 clears 60% average success rate; most models fall below 7%, and no editor exceeds 36% on the Swap-Block task even with perfect target localization. We further show that code-level editing can match the strongest pixel-level editor, revealing complementary strengths across tasks. InfoEdit identifies reflow as a central challenge in structured visual content editing and provides a diagnostic benchmark to facilitate future progress.

Problem

• infographic editing
• global layout reasoning
• reflow
• structured visual content
• benchmark

Innovation

• Infographic Editing
• Reflow
• Global Layout Reasoning
• Benchmark
• Multimodal Foundation Models