Reproduce “RefinedGAW: A Refined German Academic Web Dataset” 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

Maximilian Idahl, Oleh Astappiev, Benedikt Droste, Robert Scholz, Björn Plüster, Jan Philipp Harries, Robert Jäschke, Wolfgang Nejdl

RefinedGAW turns an archived German academic website collection into an annotated index of extracted HTML and PDF texts, retaining source and capture history. It supports targeted selection. Its academic-research category contains 87.6 billion GPT-OSS tokens, versus 64.9 billion in FineWeb2-DE and FinePDFs-DE combined; this annotation-based result does not establish model gains, and its PDF yield is lower than FinePDFs-DE.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.