针对多作物病害识别，提出硬路由专家混合模型，实现低成本增量更新与稳定拒识。
AdapterMoE: A Two-Stage Hard-Routing Mixture-of-Experts Architecture for Multi-Crop Disease Recognition with Calibrated Rejection and Incremental Learning
- 采用硬路由机制，按作物分类并拒绝非目标作物输入。
- 训练成本降至基线的9%，新作物扩展仅需局部更新。
- 支持增量学习与分布外检测，适合实际部署场景。
及时识别作物病害对粮食安全至关重要。多作物识别适合使用专家混合（MoE）架构，但传统软路由MoE端到端自由分配作物，导致少数专家主导（专家崩溃），且无语义对应关系，面临重训练成本高、非目标输入拒识不稳定及准确率饱和等问题。本文将目标从精度转向部署成本、可扩展性与拒识稳定性之间的权衡，采用确定性硬路由。提出AdapterMoE：RouterHead通过最大软概率阈值分类作物并拒识非目标作物，结合能量+KNN双门控分布外检测模块捕捉分布偏移输入；五个基于冻结EfficientNet-B0主干的每作物适配器（Adapter）分别判别病害，各通过温度缩放校准。由于专家在数据层面硬隔离，设计避免了专家崩溃，并提供add_crop接口实现局部、按作物更新，无需全量重训。在PlantVillage数据集（5作物，26类）上，五系统公平对比中，AdapterMoE的宏F1与最优基线统计上无差异（差距在0.24点内），训练成本降至全网络基线的约9%，新作物扩展效率显著提升。
原文摘要 · Abstract (English)
Timely crop-disease identification is critical to food security. Multi-crop recognition suits Mixture-of-Experts (MoE), but conventional soft-routing MoE learns crop assignment freely end-to-end, letting a few experts dominate (expert collapse) with no semantic correspondence to crops, and facing high retraining costs, unstable rejection of non-target inputs, and a saturated accuracy ceiling. We shift the objective from accuracy toward a trade-off among deployment cost, scaling flexibility, and rejection stability, using deterministic hard routing. We propose AdapterMoE: a RouterHead classifies the crop and rejects non-target crops via a Maximum Softmax Probability threshold, with a dual-gate Energy+KNN out-of-distribution module catching distribution-shifted inputs; five per-crop Adapters atop a frozen EfficientNet-B0 backbone discriminate diseases, each calibrated via Temperature Scaling. Because experts are hard-isolated at the data level, the design avoids expert collapse and exposes an add_crop interface for local, per-crop updates instead of full retraining. On PlantVillage (5 crops, 26 classes), across a fair five-system comparison, AdapterMoE attains accuracy statistically indistinguishable from the best baselines (Macro-F1 within a 0.24-point band) while cutting training cost to about 9% of full-network baselines, expanding to a new crop in
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。