Salvato in:
Dettagli Bibliografici
Autori principali: Zhao, Haozhe, Si, Shuzheng, Wang, Zhenhailong, Wang, Zheng, Chen, Liang, Li, Xiaotong, Liang, Zhixiang, Sun, Maosong, Zhang, Minjia
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2605.30611
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913171973865472
author Zhao, Haozhe
Si, Shuzheng
Wang, Zhenhailong
Wang, Zheng
Chen, Liang
Li, Xiaotong
Liang, Zhixiang
Sun, Maosong
Zhang, Minjia
author_facet Zhao, Haozhe
Si, Shuzheng
Wang, Zhenhailong
Wang, Zheng
Chen, Liang
Li, Xiaotong
Liang, Zhixiang
Sun, Maosong
Zhang, Minjia
contents Scientific figures are among the most effective means of communicating complex research ideas, yet producing publication-quality illustrations remains one of the most labor-intensive parts of paper preparation. Existing automated systems each target a single figure type under text-only input, leaving the diversity of types and conditions researchers actually use unaddressed; their raster outputs further cannot be locally revised. Because scientific figures are structured compositions of discrete semantic components, the localized errors generators produce on such layouts demand not a stronger backbone but a harness. We instantiate this harness in two complementary systems: Crafter, a multi-agent harness for figure generation that generalizes across figure types and input conditions without architectural changes, and CraftEditor, which applies the same pattern to convert raster outputs into editable SVGs. Moreover, we introduce CraftBench, a benchmark spanning three figure types and four input conditions with human quality annotation. Experiments show that Crafter substantially outperforms both standalone generators and the agentic baseline on PaperBanana-Bench and CraftBench, with ablations confirming each component's independent contribution; CraftEditor faithfully converts outputs into editable SVGs that surpass all baselines. Our code and benchmark are available at https://github.com/HaozheZhao/Crafter.
format Preprint
id arxiv_https___arxiv_org_abs_2605_30611
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs
Zhao, Haozhe
Si, Shuzheng
Wang, Zhenhailong
Wang, Zheng
Chen, Liang
Li, Xiaotong
Liang, Zhixiang
Sun, Maosong
Zhang, Minjia
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Scientific figures are among the most effective means of communicating complex research ideas, yet producing publication-quality illustrations remains one of the most labor-intensive parts of paper preparation. Existing automated systems each target a single figure type under text-only input, leaving the diversity of types and conditions researchers actually use unaddressed; their raster outputs further cannot be locally revised. Because scientific figures are structured compositions of discrete semantic components, the localized errors generators produce on such layouts demand not a stronger backbone but a harness. We instantiate this harness in two complementary systems: Crafter, a multi-agent harness for figure generation that generalizes across figure types and input conditions without architectural changes, and CraftEditor, which applies the same pattern to convert raster outputs into editable SVGs. Moreover, we introduce CraftBench, a benchmark spanning three figure types and four input conditions with human quality annotation. Experiments show that Crafter substantially outperforms both standalone generators and the agentic baseline on PaperBanana-Bench and CraftBench, with ablations confirming each component's independent contribution; CraftEditor faithfully converts outputs into editable SVGs that surpass all baselines. Our code and benchmark are available at https://github.com/HaozheZhao/Crafter.
title Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2605.30611