cs.AIMar 24, 2026

AgriPestDatabase-v1.0: A Structured Insect Dataset for Training Agricultural Large Language Model

Authors: Yagizhan Bilal Durak, Ahsan Ul Islam, Shahidul Islam, Ashley Morgan-Olvera, Iftekhar Ibne Basith, Syed Hasib Akhter Faruqui

Organizations: Sam Houston State University, Huntsville, TX, USA · Kennesaw State University, Kennesaw, GA, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Agricultural pest management increasingly relies on timely and accurate access to expert knowledge, yet high quality labeled data and continuous expert support remain limited, particularly for farmers operating in rural regions with unstable/no internet connectivity. At the same time, the rapid growth of AI and LLMs has created new opportunities to deliver practical decision support tools directly to end users in agriculture through compact and deployable systems. This work addresses (i) generating a structured insect information dataset, and (ii) adapting a lightweight LLM model (≤\leq 7B) by fine tuning it for edge device uses in agricultural pest management. The textual data collection was done by reviewing and collecting information from available pest databases and published manuscripts on nine selected pest species. These structured reports were then reviewed and validated by a domain expert. From these reports, we constructed Q/A pairs to support model training and evaluation. A LoRA-based fine-tuning approach was applied to multiple lightweight LLMs and evaluated. Initial evaluation shows that Mistral 7B achieves an 88.9% pass rate on the domain-specific Q/A task, substantially outperforming Qwen 2.5 7B (63.9%), and LLaMA 3.1 8B (58.7%). Notably, Mistral demonstrates higher semantic alignment (embedding similarity: 0.865) despite lower lexical overlap (BLEU: 0.097), indicating that semantic understanding and robust reasoning are more predictive of task success than surface-level conformity in specialized domains. 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 and demonstrates the feasibility of deploying compact, high-performing language models for practical field-level pest management guidance.

Figures & tables

Fig. 1 : End-to-End Pipeline for Expert-Curated data collection, Text Processing, Q/A Pair Generation, and Model Fine-Tuning for Invasive Pest Management.
Fig. 2 : Harmful crop insects: a) Citrus longhorned beetle b) Redheaded pine sawfly c) Asian citrus psyllid d) Emerald ash borer e) Cotton cutworm f) Six-toothed bark beetle g) Large pine weevil h) red palm weevil
Fig. 3 : Topic modeling results: (a) network-based thematic structure highlighting inter-topic relationships; (b) quantitative distribution of topic frequencies across documents.
Fig. 4 : Q/A dataset construction and LoRA-based fine-tuning pipeline. Left : chunking, LLM-based Q/A generation, cleaning, augmentation, and train–test split. Right : LoRA adapter (PEFT), loss masking, fine-tuning, and evaluation.
MetricMistral 7BLLaMA 3.1 8BQwen2.5 7B
BLEU0.09660.20490.1823
ROUGE-10.17420.36800.2968
ROUGE-20.16180.25770.2216
ROUGE-L0.17100.33380.2741
Embedding Similarity0.86500.82050.8164
Token F10.34620.40600.3736
TABLE I : Evaluation Results Across Models ( n=2,510n=2,510 )
More in this topic

LLM Fine-Tuning

LLM fine-tuning adapts pretrained large language models to tasks, domains, languages, and behaviors through additional training.

See all 754 papers

Explore similar work

CardsList
Jun 27, 2026

Fine-Tuning General-Purpose Large Language Models for Agricultural Applications:A Reproducible Framework and Evaluation Protocol Based on Qwen3-8B

General-purpose large language models (LLMs) have demonstrated strong abilities in opendomain question answering, information extraction, and text generation. Agricultural applications, however, are domain-specific, region-dependent, time-sensitive, and safety-critical. Without data governance, expert evaluation, and evidence constraints, an agricultural assistant mayproduce unreliable advice on crop diseases, pesticide use, fertilization, or policy interpretation.To avoid presenting unverified simulated numbers as real experimental findings, this paper doesnot report any model-performance claims that have not been produced by an actual training runand expert evaluation. Instead, we propose AgriTune-R, a reproducible and auditable frameworkfor adapting general-purpose LLMs to agricultural tasks. The framework selects the publiclyverifiable Qwen3-8B model as the recommended base model and integrates agricultural datagovernance, instruction construction, LoRA/QLoRA parameter-efficient fine-tuning, retrievalaugmented generation, expert evaluation, and safety control for high-risk questions. The contributions are: (1) a structured workflow for agricultural LLM adaptation; (2) an evaluationprotocol for agricultural knowledge QA, pest and disease consultation, cultivation management,and policy explanation; (3) an expert-review rubric combining factuality, safety, evidence consistency, and uncertainty expression; and (4) a clear separation between protocol design andempirical conclusions, providing an executable baseline for future empirical studies.

Zhaoyang Li, Ruijie Zhang, Jiaqi Liu +1

1Sanya University · 2Hebei International Studies University

LLM Fine-Tuning
AskPDF
May 7, 2026

Pest-Thinker: Learning to Think and Reason like Entomologists via Reinforcement Learning

Pest-induced crop losses pose a major threat to global food security and sustainable agricultural development. While recent advances in Multimodal Large Language Models (MLLMs) have shown strong potential for visual understanding and smart agriculture, their direct application to pest recognition remains limited due to the domain's unique challenges such as high inter-species complexity, intra-species variability, and the scarcity of expert-annotated data. In this work, we introduce Pest-Thinker, a knowledge-driven reinforcement learning (RL) framework that enables MLLMs to reason over fine-grained pest morphology. We first construct two high-definition pest benchmarks, QFSD and AgriInsect, comprising diverse species and expert-annotated morphological traits. Leveraging these datasets, we synthesize Chain-of-Thought (CoT) reasoning trajectories to facilitate structured learning of pest-specific visual cues through Supervised Fine-Tuning (SFT). Subsequently, we employ Group Relative Policy Optimization (GRPO) with a novel feature reward that guides the model to focus on observable morphological evidence, assessed by an LLM-as-a-Judge strategy. Extensive experiments demonstrate that Pest-Thinker substantially improves both in-domain and out-of-domain morphological understanding, marking a step toward expert-level visual reasoning for intelligent agricultural pest analysis. The datasets and source code are available upon acceptance.

Xueheng Li, Yu Wang, Tao Hu +6

Institute of Intelligent Machines, Hefei Institute of Physical Science, Chinese Academy of Sciences · University of Science and Technology of China · 4Hefei University of Technology +1

Agricultural Image AnalysisVLM ReasoningFine-Grained Visual PerceptionRL for VLMs
AskPDF
Jun 28, 2026

KrishokChat: A Citation-Grounded Dataset and Benchmark for Bengali Agricultural Advisory

We present KrishokChat, the first citation-grounded Bengali agricultural instruction-tuning dataset for crop advisory in low-resource settings. We establish a foundation of 290 hierarchical Knowledge Nodes, extracting disease symptoms, management practices, chemical dosages, and verbatim citations from 129 domain-filtered agricultural manuals. Every training instance inherits a verified citation header, guaranteeing 100% citation provenance. Using a Partitioned Seed Generation Matrix, these nodes are expanded into 139,200 supervised fine-tuning pairs, and augmented with 5,300 chemical safety and 1,000 adversarial safety instances, yielding 145,500 QA pairs across 18 crop categories. To evaluate real-world performance, we introduce the Farmer Benchmark, comprising 1,001 authentic farmer queries curated from field surveys and digital portals. Empirical evaluation on Gemma-4-E2B reveals that while fine-tuning on KrishokChat vastly improves structured formatting, standalone models still struggle with exact chemical dosage generalization. This highlights the dataset's true value as a verified knowledge base for retrieval-augmented generation (RAG) rather than mere parametric memorization. All data, code, and benchmarks are released under CC-BY-4.0.

Khan Raiyan Ibne Reza, Omar Ibne Shahid

North South University Dhaka, Bangladesh

LLM GroundingQuestion AnsweringSynthetic Data Generation for Language ModelsKnowledge-Intensive QA
AskPDF