English

VCG-Bench: Towards A Unified Visual-Centric Benchmark for Structured Generation and Editing

Computation and Language 2026-05-18 v1 Computer Vision and Pattern Recognition

Abstract

Despite the rapid advancements in Vision-Language Models (VLMs), a critical gap remains in their ability to handle structured, controllable diagrammatic tasks essential for professional workflows. Existing methods predominantly rely on pixel-based synthesis, which operates in probabilistic pixel spaces and is inherently limited in editability and fidelity. Instead, we propose a new Diagram-as-Code paradigm with symbolic logic that leverages mxGraph Extensible Markup Language (XML) for precise diagram generation and editing. We present VCG-Bench, a unified benchmark for visual-centric \texttt{mxGraph} tasks. VCG-Bench comprises: (1) a taxonomized dataset of 1,449 diverse diagrams spanning 6 domains and 15 sub-domains, (2) a paradigm definition that integrates Generation (Vision-to-Code) and Editability (Code-to-Code), (3) a Tailored Evaluation Protocol employing multi-dimensional metrics such as \texttt{mxGraph} Execution Success Rate, Style Consistency Score (SCS), etc. Experimental results highlight the challenges faced by current State-of-the-Art (SOTA) VLMs in structured fidelity and instruction compliance, reflecting their vision and reasoning capabilities.

Keywords

Cite

@article{arxiv.2605.15677,
  title  = {VCG-Bench: Towards A Unified Visual-Centric Benchmark for Structured Generation and Editing},
  author = {Xiaoyan Su and Peijie Dong and Zhenheng Tang and Song Tang and Yuyao Zhai and Kaitao Lin and Liang Chen and Gai Yuhang and Yuyu Luo and Qiang Wang and Xiaowen Chu},
  journal= {arXiv preprint arXiv:2605.15677},
  year   = {2026}
}

Comments

Accepted by ICML2026, 37 pages, 10 figures

R2 v1 2026-07-22T07:13:50.728Z