Paper Detail

SciDiagramEdit: Learning to Edit Scientific Diagrams from Paper Revisions

Yasheng Sun, Zezi Zeng, Yifan Yang, Chong Luo, Wenyi Wang, Ziwei Liu, Jürgen Schmidhuber

arxiv Score 10.2

Published 2026-07-16 · First seen 2026-07-17

General AI

Abstract

Editing the figures in a research paper is a routine and time-consuming part of everyday research practice: authors relabel components, rearrange panels, and restyle visuals as they revise their manuscripts. Automating this editing workflow under a natural-language instruction, however, is challenging, because a scientific figure is a dense infographic in which heterogeneous visual elements such as schematics, plots, photos, captions, and arrows are composed under a tight visual grammar to advance a specific argument. To address this, we present SciDiagramEdit, a benchmark and skill-evolution framework that learns from natural paper revisions and operates on the figure's editable vector source, where users can inspect and co-edit individual primitives alongside the agent. Our benchmark mines before/after figure pairs from arXiv version histories, each grounded in the authors' own revision intent. To accommodate the diversity of editing instructions, we adopt agentic learning via skill evolution: an agentic proposer continually refines the agent's skill specification from execution traces over multiple epochs. The resulting skill progressively lifts edit accuracy on a held-out validation set, providing evidence that natural paper revisions are an effective training signal for instruction-driven figure editing.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{sun2026scidiagramedit,
  title = {SciDiagramEdit: Learning to Edit Scientific Diagrams from Paper Revisions},
  author = {Yasheng Sun and Zezi Zeng and Yifan Yang and Chong Luo and Wenyi Wang and Ziwei Liu and Jürgen Schmidhuber},
  year = {2026},
  abstract = {Editing the figures in a research paper is a routine and time-consuming part of everyday research practice: authors relabel components, rearrange panels, and restyle visuals as they revise their manuscripts. Automating this editing workflow under a natural-language instruction, however, is challenging, because a scientific figure is a dense infographic in which heterogeneous visual elements such as schematics, plots, photos, captions, and arrows are composed under a tight visual grammar to advan},
  url = {https://arxiv.org/abs/2607.15272},
  keywords = {cs.CL, cs.AI, Computer science, Workflow, Benchmark (surveying), Artificial intelligence, Natural language},
  eprint = {2607.15272},
  archiveprefix = {arXiv},
}

Metadata

{}