Jack and the BeansTALK: Towards Question Answering in Plant Biology
Keywords: Plant Biology, Knowledge Graphs, Question Answering
Abstract: Developing effective question-answering systems tailored to specific domains remains an active area of research in NLP. This paper details the first steps in creating BeansTALK, an innovative question-answering system explicitly designed for plant biology. LLMs have contributed to a rapid rise in consumer adoption of generative AI technologies; however, in specific research fields, their applications are still minimal. To address this, we propose BeansTALK, which utilizes a corpus of published articles to construct a knowledge graph via an LLM. This graph will form the core of our system, which incorporates Retrieval-Augmented Generation (RAG) to enhance answer precision and relevance. Future work will leverage BeansTALK's potential to assist with literature reviews, enhance accessibility in plant biology research, and stimulate collaborative endeavors within the discipline.
Submission Number: 35
Loading