BananaVLM: A Domain-Adapted Vision Language Model for Banana Crop Disease Diagnosis
Abstract
Banana crop diseases threaten food security across the world, yet field diagnosis remains difficult because of limited expert access and visual similarity among disease classes. We present BananaVLM, a domain-adapted vision-language model for banana disease diagnosis built on LLaVA-v1.5-7B and fine-tuned with Low-Rank Adaptation (LoRA). To address the lack of multimodal agricultural instruction data, we introduce BananaInstruct, an automated pipeline that converts raw disease images into ≈80,000 question--answer pairs without manual annotation, using LLaVA-1.5-13B for symptom descriptions and Mistral-7B for agricultural Q&A generation and label grounding. Evaluated against 14 open-source and 5 closed-source VLMs on a nine-class benchmark, BananaVLM achieves 92.21% in-domain and 83.28% out-of-domain (OOD) classification accuracy, outperforming the best closed-source baseline, Gemini~2.5~Pro, by 49.8 and 63.3 percentage points, respectively. Binary healthy/diseased identification reaches 98.38% OOD accuracy with 1.00 recall. A controlled LoRA--DoRA comparison shows that DoRA performance degrades across epochs, while LoRA remains more effective for fine-grained classification. Qualitative evaluation using nine LLM judges yields G-Eval win rates of 0.81--0.96, while five domain experts prefer BananaVLM in 98.48% of 990 blind pairwise comparisons. These results show that lightweight domain adaptation with automated instruction tuning is an effective and scalable approach for specialized agricultural AI. Code, datasets, and model weights are available at https://github.com/samy101/banana-vlm
Related
- Analysis of Quantized and Efficiently Adapted Protein Language Models
- Integrating large language model-generated culturally adapted verbal suggestions into music therapy for university students: A pilot randomized exploratory study
- A domain-general basal ganglia mechanism for inhibitory control during language comprehension
- Language-guided domain adaptation improves protein sequence rep-resentations for antimicrobial resistance analysis
- A target-domain augmentation and class-wise alignment framework for few-shot cross-condition bearing fault diagnosis
BibTeX
@misc{jena2026bananavlm,
title = {{BananaVLM: A Domain-Adapted Vision Language Model for Banana Crop Disease Diagnosis}},
author = {Sangam Kumar Jena and Pandarasamy Arjunan},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.25040},
eprint = {2609.25040},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.25040},
note = {CC BY-SA 4.0},
}
RIS
TY - UNPB
TI - BananaVLM: A Domain-Adapted Vision Language Model for Banana Crop Disease Diagnosis
AU - Sangam Kumar Jena
AU - Pandarasamy Arjunan
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2609.25040
UR - https://arxiv.org/abs/2609.25040
AB - Banana crop diseases threaten food security across the world, yet field diagnosis remains difficult because of limited expert access and visual similarity among disease classes. We present BananaVLM, a domain-adapted vision-language model for banana disease diagnosis built on LLaVA-v1.5-7B and fine-tuned with Low-Rank Adaptation (LoRA). To address the lack of multimodal agricultural instruction data, we introduce BananaInstruct, an automated pipeline that converts raw disease images into ≈80,000 question--answer pairs without manual annotation, using LLaVA-1.5-13B for symptom descriptions and Mistral-7B for agricultural Q&A generation and label grounding. Evaluated against 14 open-source and 5 closed-source VLMs on a nine-class benchmark, BananaVLM achieves 92.21% in-domain and 83.28% out-of-domain (OOD) classification accuracy, outperforming the best closed-source baseline, Gemini~2.5~Pro, by 49.8 and 63.3 percentage points, respectively. Binary healthy/diseased identification reaches 98.38% OOD accuracy with 1.00 recall. A controlled LoRA--DoRA comparison shows that DoRA performance degrades across epochs, while LoRA remains more effective for fine-grained classification. Qualitative evaluation using nine LLM judges yields G-Eval win rates of 0.81--0.96, while five domain experts prefer BananaVLM in 98.48% of 990 blind pairwise comparisons. These results show that lightweight domain adaptation with automated instruction tuning is an effective and scalable approach for specialized agricultural AI. Code, datasets, and model weights are available at https://github.com/samy101/banana-vlm
ER -