Metraipolis › archive › dc_e1eb35248ac0
EdiTikZ: Scientific Figure Editing from Revision Trajectories — arXiv:2609.01409
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.01409
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T06:32:28.852Z
- SHA-256 of the archived text
b4c86a0757d9c0b89e678e445fa1058ab126f03017b8c91656ff1f6b2b5a1b3f- Stable URL of this page
https://metraipolis.com/documento/dc_e1eb35248ac0
How to cite
EdiTikZ: Scientific Figure Editing from Revision Trajectories — arXiv:2609.01409 — arXiv (en), https://arxiv.org/abs/2609.01409 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e1eb35248ac0 (Metraipolis archive dc_e1eb35248ac0, sha256 b4c86a0757d9c0b89e678e445fa1058ab126f03017b8c91656ff1f6b2b5a1b3f)
BibTeX
@misc{dce1eb35248ac0,
title = {EdiTikZ: Scientific Figure Editing from Revision Trajectories — arXiv:2609.01409},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.01409},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e1eb35248ac0; sha256 b4c86a0757d9c0b89e678e445fa1058ab126f03017b8c91656ff1f6b2b5a1b3f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01409. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language models (VLMs) have shown strong performance in generating scientific figures from text or images. However, publication-ready figures often require iterative refinement, making scientific figure editing an important yet largely unexplored step toward interactive figure creation. Existing approaches rely on costly proprietary agentic systems, focus primarily on evaluation, or construct training supervision from synthetically generated edits. Instead, we leverage naturally occurring scientific revision and development trajectories as a scalable source of supervision. To this end, we introduce DaEdiTikZ, the first large-scale dataset of revision-derived scientific figure edits, constructed by mining 391K plausible TikZ edit pairs from arXiv, GitHub, and TeX SE and inferring 781K directed edit instructions with a VLM conditioned on rendered figures and TikZ code. We further introduce DaEdiTikZ-Bench, a human-refined benchmark with 690 instances, and train two compact Qwen3.5-based EdiTikZ models (4B and 9B) by jointly learning image-to-TikZ reconstruction and instruction-conditioned editing, followed by reinforcement learning (RL) with complementary rewards for rendered fidelity and edit application. Automatic evaluation places our 9B model above all tested baselines, while human evaluation with 9 annotators and 4,320 ratings places it above GPT-5.6-Sol and on par with Gemini-3.1-Pro. Under severe out-of-distribution shifts, it remains competitive with GPT-5.6-Sol near its 2K training sequence-length regime. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.01409 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T06:15:12.695Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).