Reproduce “Hierarchical Multi-Label Classification of 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
Mobashir Sadat, Cornelia Caragea
The authors introduce SciHTC, a resource of 186,160 ACM papers and 1,233 hierarchical topics. They evaluate models on 83 topics at levels one and two, using hierarchical classification and a multi-task keyword-labeling method that does not require author keywords at test time. This method scores 33.78 Macro-F1, versus 32.67 for keyword-appended training; results cover neither full hierarchy nor other fields.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.