Reproduce “PaperFit: Vision-in-the-Loop Typesetting Optimization for Scientific Documents” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Bihui Yu, Xinglong Xu, Junjie Jiang, Jiabei Cheng, Caijun Jia, Siyuan Li, Conghui He, Jingxuan Wei and 1 more
PaperFit repairs manuscript layouts by combining source, compile logs, PDFs, and page images to diagnose defects, apply constrained edits, and recheck rendered documents. On PaperFit-Bench, it scores 3.3907 versus VisualMR’s 2.8006 on gated visual quality and hits exact page budgets in 0.8050 versus 0.5487 of cases. Results establish this advantage on a benchmark of 200 instances, not all submission workflows.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.