Saved in:
Bibliographic Details
Main Authors: Wang, Tianfu, Ding, Leilei, Tao, Ziyang, Zhan, Yi, Ma, Zhiyuan, Wu, Wei, Lei, Yuxuan, Feng, Yuan, Wang, Junyang, Wu, Yin, Xu, Yizhao, Zhu, Hongyuan, Liu, Qi, Yuan, Nicholas Jing, Zhang, Yanyong, Xiong, Hui
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.09568
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917399906746368
author Wang, Tianfu
Ding, Leilei
Tao, Ziyang
Zhan, Yi
Ma, Zhiyuan
Wu, Wei
Lei, Yuxuan
Feng, Yuan
Wang, Junyang
Wu, Yin
Xu, Yizhao
Zhu, Hongyuan
Liu, Qi
Yuan, Nicholas Jing
Zhang, Yanyong
Xiong, Hui
author_facet Wang, Tianfu
Ding, Leilei
Tao, Ziyang
Zhan, Yi
Ma, Zhiyuan
Wu, Wei
Lei, Yuxuan
Feng, Yuan
Wang, Junyang
Wu, Yin
Xu, Yizhao
Zhu, Hongyuan
Liu, Qi
Yuan, Nicholas Jing
Zhang, Yanyong
Xiong, Hui
contents High-fidelity diagram creation requires the complex orchestration of semantic topology, visual styling, and spatial layout, posing a significant challenge for automated systems. Existing methods also suffer from a representation gap: pixel-based models often lack precise control, while code-based synthesis limits intuitive flexibility. To bridge this gap, we introduce EvoDiagram, an agentic framework that generates object-level editable diagrams via an intermediate canvas schema. EvoDiagram employs a coordinated multi-agent system to decouple semantic intent from rendering logic, resolving conflicts across heterogeneous design layers. Additionally, we propose a design knowledge evolution mechanism that distills execution traces into a hierarchical memory of domain guidelines, enabling agents to retrieve context-aware expertise adaptively. We further release CanvasBench, a benchmark consisting of both data and metrics for canvas-based diagramming. Extensive experiments demonstrate that EvoDiagram exhibits excellent performance and balance against baselines in generating editable, structurally consistent, and aesthetically coherent diagrams. Our code is available at https://github.com/AuraX-AI/EvoDiagram.
format Preprint
id arxiv_https___arxiv_org_abs_2604_09568
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EvoDiagram: Agentic Editable Diagram Creation via Design Expertise Evolution
Wang, Tianfu
Ding, Leilei
Tao, Ziyang
Zhan, Yi
Ma, Zhiyuan
Wu, Wei
Lei, Yuxuan
Feng, Yuan
Wang, Junyang
Wu, Yin
Xu, Yizhao
Zhu, Hongyuan
Liu, Qi
Yuan, Nicholas Jing
Zhang, Yanyong
Xiong, Hui
Human-Computer Interaction
Computation and Language
Computer Vision and Pattern Recognition
High-fidelity diagram creation requires the complex orchestration of semantic topology, visual styling, and spatial layout, posing a significant challenge for automated systems. Existing methods also suffer from a representation gap: pixel-based models often lack precise control, while code-based synthesis limits intuitive flexibility. To bridge this gap, we introduce EvoDiagram, an agentic framework that generates object-level editable diagrams via an intermediate canvas schema. EvoDiagram employs a coordinated multi-agent system to decouple semantic intent from rendering logic, resolving conflicts across heterogeneous design layers. Additionally, we propose a design knowledge evolution mechanism that distills execution traces into a hierarchical memory of domain guidelines, enabling agents to retrieve context-aware expertise adaptively. We further release CanvasBench, a benchmark consisting of both data and metrics for canvas-based diagramming. Extensive experiments demonstrate that EvoDiagram exhibits excellent performance and balance against baselines in generating editable, structurally consistent, and aesthetically coherent diagrams. Our code is available at https://github.com/AuraX-AI/EvoDiagram.
title EvoDiagram: Agentic Editable Diagram Creation via Design Expertise Evolution
topic Human-Computer Interaction
Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.09568