Technology Category
Machine Learning: Large Multimodal Models (LMMs)Cognitive Modeling & Cognitive Systems: Agent ArchitecturesKnowledge Representation and Reasoning: Knowledge Acquisition
AgriGPT: a Large Language Model Ecosystem for Agriculture
🤖 AI Summary
Agriculture has long suffered from the absence of domain-specific large language models (LLMs), scarcity of high-quality training data, and inadequate evaluation benchmarks—hindering LLM deployment in this critical sector. To address these challenges, we introduce the first open-source LLM ecosystem for agriculture: (1) a multi-agent data engine that synthesizes the high-quality agricultural QA dataset Agri-342K; (2) Tri-RAG, a novel three-channel retrieval-augmented generation framework integrating dense retrieval, sparse retrieval, and multi-hop knowledge graph reasoning; and (3) AgriBench-13K, a comprehensive benchmark covering 13 agricultural task categories. Experiments demonstrate substantial improvements in factual consistency, knowledge retrieval accuracy, and complex reasoning capability over general-purpose LLMs. All models, datasets, and code are publicly released to foster sustainable advancement of AI in agriculture.
Technology Category
Application Category
📝 Abstract
Despite the rapid progress of Large Language Models (LLMs), their application in agriculture remains limited due to the lack of domain-specific models, curated datasets, and robust evaluation frameworks. To address these challenges, we propose AgriGPT, a domain-specialized LLM ecosystem for agricultural usage. At its core, we design a multi-agent scalable data engine that systematically compiles credible data sources into Agri-342K, a high-quality, standardized question-answer (QA) dataset. Trained on this dataset, AgriGPT supports a broad range of agricultural stakeholders, from practitioners to policy-makers. To enhance factual grounding, we employ Tri-RAG, a three-channel Retrieval-Augmented Generation framework combining dense retrieval, sparse retrieval, and multi-hop knowledge graph reasoning, thereby improving the LLM's reasoning reliability. For comprehensive evaluation, we introduce AgriBench-13K, a benchmark suite comprising 13 tasks with varying types and complexities. Experiments demonstrate that AgriGPT significantly outperforms general-purpose LLMs on both domain adaptation and reasoning. Beyond the model itself, AgriGPT represents a modular and extensible LLM ecosystem for agriculture, comprising structured data construction, retrieval-enhanced generation, and domain-specific evaluation. This work provides a generalizable framework for developing scientific and industry-specialized LLMs. All models, datasets, and code will be released to empower agricultural communities, especially in underserved regions, and to promote open, impactful research.
Problem
Research questions and friction points this paper is trying to address.
Lack of domain-specific LLMs for agricultural applications
Absence of curated datasets and robust evaluation frameworks
Need for reliable reasoning in agricultural decision-making
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multi-agent scalable data engine for Agri-342K dataset
Tri-RAG framework enhances factual grounding
AgriBench-13K benchmark for comprehensive evaluation
🔎 Similar Papers
No similar papers found.
