CPJ: Explainable Agricultural Pest Diagnosis via Caption-Prompt-Judge with LLM-Judged Refinement

Wentao Zhang, Tao Fang, Lina Lu, Lifei Wang, Weihe Zhong
12/31/2025
cs.CVcs.CL

Official code: CPJ-Agricultural/CPJ-Agricultural-Diagnosis (4 stars, MIT, last commit 2026-01-08).

Abstract

Accurate and interpretable crop disease diagnosis is essential for agricultural decision-making, yet existing methods often rely on costly supervised fine-tuning and perform poorly under domain shifts. We propose Caption--Prompt--Judge (CPJ), a training-free few-shot framework that enhances Agri-Pest VQA through structured, interpretable image captions. CPJ employs large vision-language models to generate multi-angle captions, refined iteratively via an LLM-as-Judge module, which then inform a dual-answer VQA process for both recognition and management responses. Evaluated on CDDMBench, CPJ significantly improves performance: using GPT-5-mini captions, GPT-5-Nano achieves \textbf{+22.7} pp in disease classification and \textbf{+19.5} points in QA score over no-caption baselines. The framework provides transparent, evidence-based reasoning, advancing robust and explainable agricultural diagnosis without fine-tuning. Our code and data are publicly available at: https://github.com/CPJ-Agricultural/CPJ-Agricultural-Diagnosis.

View on arXivView PDF

Code Implementations(1)

CPJ-Agricultural/CPJ-Agricultural-DiagnosisOfficial100%
MIT
41PythonDec 19, 20259 months agoMIT

Similar Papers

Agri-R1: Empowering Generalizable Agricultural Reasoning in Vision-Language Models with Reinforcement Learning

85% match2026

Agro-Consensus: Semantic Self-Consistency in Vision-Language Models for Crop Disease Management in Developing Countries

83% match2025

AgriGPT-VL: Agricultural Vision-Language Understanding Suite

82% match2025

Are vision-language models ready to zero-shot replace supervised classification models in agriculture?

82% match2025

A Lightweight and Explainable Vision-Language Framework for Crop Disease Visual Question Answering

82% match2026

LeafNet: A Large-Scale Dataset and Comprehensive Benchmark for Foundational Vision-Language Understanding of Plant Diseases

82% match1 repo102026

Cite this paper

@article{zhang2025cpj,
  title  = {CPJ: Explainable Agricultural Pest Diagnosis via Caption-Prompt-Judge with LLM-Judged Refinement},
  author = {Wentao Zhang and Tao Fang and Lina Lu and Lifei Wang and Weihe Zhong},
  year   = {2025},
  eprint = {2512.24947},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2512.24947v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2512.24947.svg)](https://codeofpaper.com/paper/2512.24947)

Discussion