Fine-Tuning and Evaluating Conversational AI for Agricultural Advisory

• URL: http://arxiv.org/abs/2603.03294v1
• Date: Fri, 06 Feb 2026 10:18:32 GMT
• Title: Fine-Tuning and Evaluating Conversational AI for Agricultural Advisory
• Abstract summary: Large Language Models show promise for agricultural advisory, yet vanilla models exhibit unsupported recommendations.<n>We present a hybrid LLM architecture that decouples factual retrieval from conversational delivery.<n>Experiments on crops and queries from Bihar, India show that fine-tuning on curated data substantially improves fact recall and F1.
• Score: 0.5839782776991492
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Fine-Tuning General-Purpose Large Language Models for Agricultural Applications:A Reproducible Framework and Evaluation Protocol Based on Qwen3-8B [14.747665015038345]
  AgriTune-R is a framework for adapting general-purpose large language models to agricultural tasks.<n>The framework integrates agricultural data governance, instruction construction, LoRA/QLoRA parameter-efficient fine-tuning, retrievalaugmented generation, expert evaluation, and safety control.
  arXiv  Detail & Related papers  (2026-06-27T16:02:50Z)
• Trust but Verify: Introducing DAVinCI -- A Framework for Dual Attribution and Verification in Claim Inference for Language Models [53.279391576560755]
  Large Language Models (LLMs) have demonstrated remarkable fluency and versatility across a wide range of NLP tasks, yet they remain prone to factual inaccuracies and limitation.<n>This poses significant risks in high-stakes domains such as healthcare, law, and scientific communication, where trust and verifiability are paramount.<n>We introduce DAVinCI - a Dual Attribution and Verification framework designed to enhance the factual reliability and interpretability of LLM outputs.
  arXiv  Detail & Related papers  (2026-04-23T01:37:50Z)
• Co-FactChecker: A Framework for Human-AI Collaborative Claim Verification Using Large Reasoning Models [99.26398772227684]
  We propose Co-FactChecker, a framework for human-AI collaborative claim verification.<n>Co-FactChecker translates expert feedback into trace-edits that introduce targeted modifications to the trace.<n>Human evaluations show that Co-FactChecker is preferred over multi-turn dialogue.
  arXiv  Detail & Related papers  (2026-04-15T10:35:00Z)
• AgriPestDatabase-v1.0: A Structured Insect Dataset for Training Agricultural Large Language Model [0.0]
  Agricultural pest management increasingly relies on timely and accurate access to expert knowledge, yet high quality labeled data and continuous expert support remain limited.<n>This work addresses (i) generating a structured insect information dataset, and (ii) adapting a lightweight LLM model by fine tuning it for edge device uses in agricultural pest management.<n>By combining expert organized data, well-structured Q/A pairs, semantic quality control, and efficient model adaptation, this work contributes towards providing support for farmer facing agricultural decision support tools.
  arXiv  Detail & Related papers  (2026-03-24T04:11:27Z)
• AgriChat: A Multimodal Large Language Model for Agriculture Image Understanding [13.665861251747144]
  We present AgriChat, a specialized MLLM that presents broad knowledge across thousands of agricultural classes and provides detailed agricultural assessments with explanations.<n>The results validate that preserving visual detail combined with web-verified knowledge constitutes a reliable pathway toward robust and trustworthy agricultural AI.
  arXiv  Detail & Related papers  (2026-03-14T11:49:05Z)
• Agri-R1: Empowering Generalizable Agricultural Reasoning in Vision-Language Models with Reinforcement Learning [22.34625628938106]
  We propose textbfAgri-R1, a reasoning-enhanced large model for agriculture.<n>Our framework high-quality reasoning data generation via vision-language synthesis and LLM-based filtering.<n>We show a +23.2% relative gain in disease recognition accuracy, +33.3% in agricultural knowledge QA, and a +26.10-point improvement in cross-domain generalization.
  arXiv  Detail & Related papers  (2026-01-08T07:34:37Z)
• Dynamic Orchestration of Multi-Agent System for Real-World Multi-Image Agricultural VQA [14.881994273322627]
  Agricultural visual question answering is essential for providing farmers and researchers with accurate and timely knowledge.<n>We propose a self-reflective and self-improving multi-agent framework that integrates four roles, the Retriever, the Reflector, the Answerer, and the Improver.<n> Experiments on the AgMMU benchmark show that our framework achieves competitive performance on multi-image agricultural QA.
  arXiv  Detail & Related papers  (2025-09-29T06:52:10Z)
• Leveraging Synthetic Data for Question Answering with Multilingual LLMs in the Agricultural Domain [1.0144032120138065]
  This study generates multilingual (English, Hindi, Punjabi) synthetic datasets from agriculture-specific documents from India.<n> Evaluation on human-created datasets demonstrates significant improvements in factuality, relevance, and agricultural consensus.
  arXiv  Detail & Related papers  (2025-07-22T19:25:10Z)
• Attributing Response to Context: A Jensen-Shannon Divergence Driven Mechanistic Study of Context Attribution in Retrieval-Augmented Generation [52.3707788779464]
  We introduce a novel Jensen-Shannon Divergence driven method to Attribute Response to Context (ARC-JSD)<n>ARC-JSD enables efficient and accurate identification of essential context sentences without additional fine-tuning, gradient-calculation or surrogate modelling.<n> Evaluations on a wide range of RAG benchmarks, such as TyDi QA, Hotpot QA, and Musique, using instruction-tuned LLMs in different scales demonstrate superior accuracy and significant computational efficiency improvements.
  arXiv  Detail & Related papers  (2025-05-22T09:04:03Z)
• AgroGPT: Efficient Agricultural Vision-Language Model with Expert Tuning [30.034193330398292]
  We propose an approach to construct instruction-tuning data that harnesses vision-only data for the agriculture domain.<n>We utilize diverse agricultural datasets spanning multiple domains, curate class-specific information, and employ large language models (LLMs) to construct an expert-tuning set.<n>We expert-tuned and created AgroGPT, an efficient LMM that can hold complex agriculture-related conversations and provide useful insights.
  arXiv  Detail & Related papers  (2024-10-10T22:38:26Z)
• Contrastive Learning to Improve Retrieval for Real-world Fact Checking [84.57583869042791]
  We present Contrastive Fact-Checking Reranker (CFR), an improved retriever for fact-checking complex claims. We leverage the AVeriTeC dataset, which annotates subquestions for claims with human written answers from evidence documents. We find a 6% improvement in veracity classification accuracy on the dataset.
  arXiv  Detail & Related papers  (2024-10-07T00:09:50Z)
• SFR-RAG: Towards Contextually Faithful LLMs [57.666165819196486]
  Retrieval Augmented Generation (RAG) is a paradigm that integrates external contextual information with large language models (LLMs) to enhance factual accuracy and relevance. We introduce SFR-RAG, a small LLM that is instruction-textual with an emphasis on context-grounded generation and hallucination. We also present ConBench, a new evaluation framework compiling multiple popular and diverse RAG benchmarks.
  arXiv  Detail & Related papers  (2024-09-16T01:08:18Z)
• FactCHD: Benchmarking Fact-Conflicting Hallucination Detection [64.4610684475899]
  FactCHD is a benchmark designed for the detection of fact-conflicting hallucinations from LLMs. FactCHD features a diverse dataset that spans various factuality patterns, including vanilla, multi-hop, comparison, and set operation. We introduce Truth-Triangulator that synthesizes reflective considerations by tool-enhanced ChatGPT and LoRA-tuning based on Llama2.
  arXiv  Detail & Related papers  (2023-10-18T16:27:49Z)
• WiCE: Real-World Entailment for Claims in Wikipedia [63.234352061821625]
  We propose WiCE, a new fine-grained textual entailment dataset built on natural claim and evidence pairs extracted from Wikipedia. In addition to standard claim-level entailment, WiCE provides entailment judgments over sub-sentence units of the claim. We show that real claims in our dataset involve challenging verification and retrieval problems that existing models fail to address.
  arXiv  Detail & Related papers  (2023-03-02T17:45:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.