PharmkGPT: Data and Evaluation Resources for a KG-Enhanced RAG Framework for Reliable Biomedical Question Answering in Alzheimer’s and Delirium
Abstract
This dataset contains the data resources and evaluation outputs supporting the study “PharmkGPT: A KG-Enhanced RAG Framework for Reliable Biomedical Question Answering in Alzheimer’s and Delirium”. PharmkGPT is a knowledge graph (KG)-enhanced retrieval-augmented generation (RAG) framework developed for evidence-grounded biomedical literature exploration and question answering, with a focus on Alzheimer’s disease and delirium.The deposited resources support the principal benchmark and supplementary analyses reported in the study. They include: (1) the internal expert-reviewed benchmark questions, answer labels and model outputs; (2) the curated retrieval corpus and question–source mappings, comprising 1,041 unique source PubMed records and 4,018 unique non-source records; (3) human-adjudicated KG entity and relation extraction outputs; (4) component-ablation retrieval and answer-generation outputs; (5) PubMedQA PQA-L retrieval and question-answering evaluation outputs; (6) blinded free-form evidence-grounded question-answering evaluation data; (7) the external gene–phenotype association benchmark derived from frozen ClinVar, PanelApp and Human Phenotype Ontology source snapshots; and (8) scalability, incremental-construction and conflicting-relation evaluation logs.The curated internal benchmark contains 1,045 questions mapped to 1,041 distinct source PubMed records and evaluates both answer correctness and exact-source retrieval. Additional evaluations assess…
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.