Leveraging Synthetic Data for Question Answering with Multilingual LLMs in the Agricultural Domain
This work addresses the need for precise, localized agricultural advice for farmers in India, though it is incremental as it applies existing fine-tuning methods to a new domain and languages.
The study tackled the problem of providing accurate, multilingual agricultural information to farmers by generating synthetic datasets in English, Hindi, and Punjabi from Indian agriculture documents and fine-tuning LLMs for question answering, resulting in significant improvements in factuality, relevance, and agricultural consensus compared to baseline models.
Enabling farmers to access accurate agriculture-related information in their native languages in a timely manner is crucial for the success of the agriculture field. Publicly available general-purpose Large Language Models (LLMs) typically offer generic agriculture advisories, lacking precision in local and multilingual contexts. Our study addresses this limitation by generating multilingual (English, Hindi, Punjabi) synthetic datasets from agriculture-specific documents from India and fine-tuning LLMs for the task of question answering (QA). Evaluation on human-created datasets demonstrates significant improvements in factuality, relevance, and agricultural consensus for the fine-tuned LLMs compared to the baseline counterparts.