Researchers develop HUGO framework to extract cold spray data from literature

Researchers have developed HUGO-CS, a novel dataset containing 4,383 cold-spray experiments extracted from scientific literature. This dataset significantly expands upon previous efforts, being over 30 times larger than the prior largest dataset. To create HUGO-CS, a framework called HUGO was employed, which combines automated LLM-based labeling with manual refinement to ensure accuracy and efficiency in data extraction from complex experimental results. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT This dataset could accelerate research and optimization in cold spray manufacturing by providing a large, structured source of experimental data.

RANK_REASON This is a research paper detailing a new dataset and extraction framework. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• HUGO-CS
• HUGO
• LLM

• paper
• other