AgroGPT: Efficient Agricultural Vision-Language Model with Expert Tuning

TL;DR

AgroGPT is a specialized vision-language model for agriculture, built using expert-tuning data from agricultural datasets and large language models, enabling complex agricultural conversations and insights.

Contribution

The paper introduces AgroInstruct, a large agricultural instruction dataset, and AgroGPT, an efficient domain-specific multimodal model trained with expert tuning, addressing the lack of domain-specific vision-language data.

Findings

AgroGPT effectively identifies fine-grained agricultural concepts.

It can act as an agricultural expert for complex questions.

Performs well compared to other open and closed models.

Abstract

Significant progress has been made in advancing large multimodal conversational models (LMMs), capitalizing on vast repositories of image-text data available online. Despite this progress, these models often encounter substantial domain gaps, hindering their ability to engage in complex conversations across new domains. Recent efforts have aimed to mitigate this issue, albeit relying on domain-specific image-text data to curate instruction-tuning data. However, many domains, such as agriculture, lack such vision-language data. In this work, we propose an approach to construct instruction-tuning data that harnesses vision-only data for the agriculture domain. We utilize diverse agricultural datasets spanning multiple domains, curate class-specific information, and employ large language models (LLMs) to construct an expert-tuning set, resulting in a 70k expert-tuning dataset called AgroInstruct. Subsequently, we expert-tuned and created AgroGPT, an efficient LMM that can hold complex agriculture-related conversations and provide useful insights. We also develop AgroEvals for evaluation and compare {AgroGPT's} performance with large open and closed-source models. {AgroGPT} excels at identifying fine-grained agricultural concepts, can act as an agriculture expert, and provides helpful information for multimodal agriculture questions. The code, datasets, and models are available at https://github.com/awaisrauf/agroGPT.