CoffeeRAG-SLM: evaluation data for on-device retrieval-augmented generation in Indonesian coffee agronomy question answering
Published: 30 September 2026| Version 1 | DOI: 10.17632/xf46ph69z8.1
Contributors:
Yudha Alif Auliya, , , , , Description
It contains every per-query record needed to reproduce the results reported in the article: the expert-validated gold-standard question–answer set, the answers generated under three system configurations, retrieval outputs, answer- and retrieval-quality scores, hallucination labels, and 900 latency/memory measurements. A script that reproduces all inferential statistics is included.
Files
Categories
Artificial Intelligence, Information Retrieval, Natural Language Processing, Coffee, Agriculture