Enhancing Agricultural Advisory Services with Multilingual LLaMA and RAG
G Bharathi Mohan, C M Jayanth Adhitya, A. Mithilesh · 2025
Agriculture, being a vital component of global food security, necessitates the development of innovative solutions to address the knowledge gap experienced by numerous farmers, particularly in developing countries where access to expert advice is scarce. In these areas, agricultural producers often depend on helplines for essential guidance, but the exorbitant costs and limited accessibility of these services create substantial obstacles. By automating responses to agricultural inquiries, the burden on traditional helpline systems can be alleviated, enabling farmers to access prompt and precise information. The combination of artificial intelligence and agriculture offers a chance to tackle these challenges, with advanced language models, especially transformers, demonstrating significant potential in comprehending intricate agricultural queries and delivering appropriate responses. This paper investigates how large language models (llms) can simplify the process of finding answers for farmers by utilizing their advanced language processing abilities. By analyzing a vast collection of over four million farmer inquiries from tamil nadu, india, encompassing diverse agricultural scenarios, this study showcases the efficacy of llms in bridging information gaps and equipping farmers with immediate access to crucial knowledge.