Saved in:
Bibliographic Details
Main Authors: Singh, Sanyam, Ganesh, Naga, Singh, Vineet, Pedapudi, Lakshmi, Kumar, Ritesh, Jyothi, SSP, Karanam, Archana, Pasha, Waseem, Kumari, Ekta, Yashoda, C., Reddy, Mettu Vijaya Rekha, Debbesa, Shesha Phani, Dash, Chandan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.03294
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908869928681472
author Singh, Sanyam
Ganesh, Naga
Singh, Vineet
Pedapudi, Lakshmi
Kumar, Ritesh
Jyothi, SSP
Karanam, Archana
Pasha, Waseem
Kumari, Ekta
Yashoda, C.
Reddy, Mettu Vijaya Rekha
Debbesa, Shesha Phani
Dash, Chandan
author_facet Singh, Sanyam
Ganesh, Naga
Singh, Vineet
Pedapudi, Lakshmi
Kumar, Ritesh
Jyothi, SSP
Karanam, Archana
Pasha, Waseem
Kumari, Ekta
Yashoda, C.
Reddy, Mettu Vijaya Rekha
Debbesa, Shesha Phani
Dash, Chandan
contents Large Language Models show promise for agricultural advisory, yet vanilla models exhibit unsupported recommendations, generic advice lacking specific, actionable detail, and communication styles misaligned with smallholder farmer needs. In high stakes agricultural contexts, where recommendation accuracy has direct consequences for farmer outcomes, these limitations pose challenges for responsible deployment. We present a hybrid LLM architecture that decouples factual retrieval from conversational delivery: supervised fine-tuning with LoRA on expert-curated GOLDEN FACTS (atomic, verified units of agricultural knowledge) optimizes fact recall, while a separate stitching layer transforms retrieved facts into culturally appropriate, safety-aware responses. Our evaluation framework, DG-EVAL, performs atomic fact verification (measuring recall, precision, and contradiction detection) against expert-curated ground truth rather than Wikipedia or retrieved documents. Experiments across multiple model configurations on crops and queries from Bihar, India show that fine-tuning on curated data substantially improves fact recall and F1, while maintaining high relevance. Using a fine-tuned smaller model achieves comparable or better factual quality at a fraction of the cost of frontier models. A stitching layer further improves safety subscores while maintaining high conversational quality. We release the farmerchat-prompts library to enable reproducible development of domain-specific agricultural AI.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03294
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Fine-Tuning and Evaluating Conversational AI for Agricultural Advisory
Singh, Sanyam
Ganesh, Naga
Singh, Vineet
Pedapudi, Lakshmi
Kumar, Ritesh
Jyothi, SSP
Karanam, Archana
Pasha, Waseem
Kumari, Ekta
Yashoda, C.
Reddy, Mettu Vijaya Rekha
Debbesa, Shesha Phani
Dash, Chandan
Computation and Language
Artificial Intelligence
Machine Learning
Large Language Models show promise for agricultural advisory, yet vanilla models exhibit unsupported recommendations, generic advice lacking specific, actionable detail, and communication styles misaligned with smallholder farmer needs. In high stakes agricultural contexts, where recommendation accuracy has direct consequences for farmer outcomes, these limitations pose challenges for responsible deployment. We present a hybrid LLM architecture that decouples factual retrieval from conversational delivery: supervised fine-tuning with LoRA on expert-curated GOLDEN FACTS (atomic, verified units of agricultural knowledge) optimizes fact recall, while a separate stitching layer transforms retrieved facts into culturally appropriate, safety-aware responses. Our evaluation framework, DG-EVAL, performs atomic fact verification (measuring recall, precision, and contradiction detection) against expert-curated ground truth rather than Wikipedia or retrieved documents. Experiments across multiple model configurations on crops and queries from Bihar, India show that fine-tuning on curated data substantially improves fact recall and F1, while maintaining high relevance. Using a fine-tuned smaller model achieves comparable or better factual quality at a fraction of the cost of frontier models. A stitching layer further improves safety subscores while maintaining high conversational quality. We release the farmerchat-prompts library to enable reproducible development of domain-specific agricultural AI.
title Fine-Tuning and Evaluating Conversational AI for Agricultural Advisory
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2603.03294