ChartSync: A Benchmark for Visuo-Logical Cascading Chart Editing

2026-07-11

AI Summary

This study addresses the challenge that generative image editing models struggle to synchronize geometric structures when modifying data in statistical charts, particularly lacking the ability to perform cascading edits that respect logical dependencies among visual elements. The work formalizes, for the first time, the “Visual-Logical Cascaded Editing” (VLCE) task and introduces ChartSync—a benchmark dataset comprising 235 procedurally generated and expert-validated chart instances with deterministic visual-logical coupling. A two-tier evaluation framework combining objective metrics and judgments from vision-language models is proposed. Evaluation across 14 image editing models and one code-based pipeline reveals that only two state-of-the-art closed-source models exhibit preliminary VLCE capability, while most open-source models suffer significant performance degradation due to semantic isolation and background distortion.

Abstract

Generative image editing models struggle with structured statistical charts when data modifications require geometric synchronization. We formalize this task as Visuo-Logical Cascading Editing (VLCE). However, existing methods remain confined to localized text substitutions and struggle with dependency-aware cascading updates. To systematically evaluate this capability, we introduce ChartSync, an expert-validated benchmark constructed via a programmatic rendering pipeline that guarantees deterministic visuo-logical coupling for the ground truth. ChartSync comprises 870 triplets across 9 chart categories and 4 task types, including 235 geometry-coupled VLCE instances that specifically test cascading text-to-geometry synchronization. We further evaluate these instances via a two-tier framework combining objective visual metrics with a vision-language model judge paradigm to assess low-level fidelity alongside multimodal comprehension and reasoning. Evaluating 14 image editing models and one code-mediated pipeline reveals a nuanced capability gap: most open-source models suffer severe drops in geometric synchronization, while only two frontier proprietary models show emerging VLCE capability, with their residual errors mainly involving semantic isolation and background corruption. Our detailed error analysis deconstructs these failure paradigms to identify core meta-abilities for guiding future multimodal architectures. The ChartSync dataset and code are publicly released at https://github.com/kaka-yjk/ChartSyncCodebase.

Problem

• visuo-logical cascading editing
• chart editing
• geometric synchronization
• multimodal reasoning
• structured data visualization

Innovation

• Visuo-Logical Cascading Editing
• ChartSync
• geometric synchronization
• multimodal reasoning
• programmatic rendering