Reproduce “Automated Keyword Generation Using LLMs in a High-Volume Data Virtualization Service” 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
Saiful Khan, Yiwen Xing, Joyraj Chakraborty, Afrah Farea, Philip Beaucamp, Samuel Jackson, Min Chen
The keyword-generation service generates searchable tags for virtual datasets using descriptions and, when available, transformation and parent-dataset context. Across 38 datasets in five workflows, Gemma3 27B reached semantic soft-F1 of 0.749, versus 0.705 for YAKE. In a metadata-only comparison, parent keywords improved scores over parent README text; these similarity-based results do not establish universal gains or real-world retrieval success.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.