BioHiCL model enhances biomedical retrieval using hierarchical contrastive learning

Researchers have developed BioHiCL, a novel approach to biomedical information retrieval that utilizes hierarchical multi-label contrastive learning. This method leverages the structured supervision from Medical Subject Headings (MeSH) to better capture semantic relationships and overlap in biomedical texts. The BioHiCL models, available in Base (0.1B parameters) and Large (0.3B parameters) versions, demonstrate strong performance on retrieval, sentence similarity, and question answering tasks while maintaining computational efficiency. AI

IMPACT Enhances biomedical information retrieval, potentially improving access to and understanding of medical research.

RANK_REASON Research paper detailing a new model and methodology. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• BioHiCL
• BioHiCL-Base
• BioHiCL-Large
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• Lecheng Zheng
• Medical Subject Headings
• ScienceCast

• paper
• model release
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →