PaperBanana: Automating Academic Illustration for AI Scientists

Dawei Zhu, Rui Meng, Yale Song, Xiyu Wei, Sujian Li, Tomas Pfister, Jinsung Yoon
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:167480-167523, 2026.

Abstract

Despite rapid advances in autonomous AI scientists powered by language models, generating publication-ready illustrations remains a labor-intensive bottleneck in the research workflow. To lift this burden, we introduce PaperBanana, an agentic framework for automated generation of publication-ready academic illustrations. Powered by state-of-the-art VLMs and image generation models, PaperBanana orchestrates specialized agents to retrieve references, plan content and style, render images, and iteratively refine via self-critique. To rigorously evaluate our framework, we introduce PaperBananaBench, comprising 292 test cases for methodology diagrams curated from NeurIPS 2025 publications, covering diverse research domains and illustration styles. Comprehensive experiments demonstrate that PaperBanana consistently outperforms leading baselines in faithfulness, conciseness, readability, and aesthetics. We further show that our method effectively extends to the generation of high-quality statistical plots. Collectively, PaperBanana paves the way for the automated generation of publication-ready illustrations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-zhu26ag, title = {{P}aper{B}anana: Automating Academic Illustration for {AI} Scientists}, author = {Zhu, Dawei and Meng, Rui and Song, Yale and Wei, Xiyu and Li, Sujian and Pfister, Tomas and Yoon, Jinsung}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {167480--167523}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/zhu26ag/zhu26ag.pdf}, url = {https://proceedings.mlr.press/v306/zhu26ag.html}, abstract = {Despite rapid advances in autonomous AI scientists powered by language models, generating publication-ready illustrations remains a labor-intensive bottleneck in the research workflow. To lift this burden, we introduce PaperBanana, an agentic framework for automated generation of publication-ready academic illustrations. Powered by state-of-the-art VLMs and image generation models, PaperBanana orchestrates specialized agents to retrieve references, plan content and style, render images, and iteratively refine via self-critique. To rigorously evaluate our framework, we introduce PaperBananaBench, comprising 292 test cases for methodology diagrams curated from NeurIPS 2025 publications, covering diverse research domains and illustration styles. Comprehensive experiments demonstrate that PaperBanana consistently outperforms leading baselines in faithfulness, conciseness, readability, and aesthetics. We further show that our method effectively extends to the generation of high-quality statistical plots. Collectively, PaperBanana paves the way for the automated generation of publication-ready illustrations.} }
Endnote
%0 Conference Paper %T PaperBanana: Automating Academic Illustration for AI Scientists %A Dawei Zhu %A Rui Meng %A Yale Song %A Xiyu Wei %A Sujian Li %A Tomas Pfister %A Jinsung Yoon %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-zhu26ag %I PMLR %P 167480--167523 %U https://proceedings.mlr.press/v306/zhu26ag.html %V 306 %X Despite rapid advances in autonomous AI scientists powered by language models, generating publication-ready illustrations remains a labor-intensive bottleneck in the research workflow. To lift this burden, we introduce PaperBanana, an agentic framework for automated generation of publication-ready academic illustrations. Powered by state-of-the-art VLMs and image generation models, PaperBanana orchestrates specialized agents to retrieve references, plan content and style, render images, and iteratively refine via self-critique. To rigorously evaluate our framework, we introduce PaperBananaBench, comprising 292 test cases for methodology diagrams curated from NeurIPS 2025 publications, covering diverse research domains and illustration styles. Comprehensive experiments demonstrate that PaperBanana consistently outperforms leading baselines in faithfulness, conciseness, readability, and aesthetics. We further show that our method effectively extends to the generation of high-quality statistical plots. Collectively, PaperBanana paves the way for the automated generation of publication-ready illustrations.
APA
Zhu, D., Meng, R., Song, Y., Wei, X., Li, S., Pfister, T. & Yoon, J.. (2026). PaperBanana: Automating Academic Illustration for AI Scientists. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:167480-167523 Available from https://proceedings.mlr.press/v306/zhu26ag.html.

Related Material