ST-LoRA: Single Trajectory LoRA Ensemble for Uncertainty Aware Agricultural Segmentation
提出ST-LoRA方法，在农业语义分割中实现高效不确定性估计，方法新颖且实验扎实，对农业AI落地有实际价值。
2026年8月2日arXiv上线cs.CV预印本ST-LoRA,提出面向农业不确定性感知语义分割的LoRA高效集成框架。研究针对数字农业决策支持对预测精度与不确定性校准的双重需求,以及集成方法计算与内存开销大的瓶颈,提出在单训练轨迹下构建多样化集成成员:每个成员共享冻结预训练骨干网络,仅通过轻量级低秩适配器(LoRA)差异,训练参数控制在全模型的10%以内,保持集成多样性。实验采用两套农业数据集:GrowliFlower-L(花椰菜,露天)和BUP20(甜椒,温室),覆盖分布内性能、分布偏移下的校准以及分布外检测。研究发现,前馈层而非注意力层是密集预测场景下LoRA的关键靶点(颠覆语言模型仅关注注意力的惯例),ST-LoRA在分割精度与校准上匹配或超越全秩集成,且显著降低训练时间、推理延迟、内存占用与存储需求;优于快照集成、MC Dropout与深度确定性不确定性等高效基线。
Abstract:Reliable decision-support in digital agriculture requires accurate predictions and well-calibrated uncertainty estimates, particularly for dense prediction tasks such as semantic segmentation. Ensemble methods provide strong uncertainty quantification, but their computational and memory demands limit practical use, while single-model approximations often trade off uncertainty quality for efficiency. We propose ST-LoRA, a parameter-efficient ensemble framework that builds diverse ensemble members from a single training trajectory by combining Low-Rank Adaptation (LoRA) with snapshot ensembling. Each member shares a frozen pretrained backbone and differs only in lightweight low-rank adapters, reducing trainable parameters to under 10% of the full model while preserving ensemble diversity. We evaluate across two agricultural datasets - GrowliFlower-L (cauliflower, open field) and BUP20 (sweet pepper, glasshouse) - using SegFormer and Mask2Former, covering in-distribution performance, calibration under distribution shift, and out-of-distribution detection. Ablations show feed-forward layers, not attention layers, are the critical LoRA target for dense prediction, contrary to the attention-only convention from language models. ST-LoRA matches or exceeds full-rank ensembles in segmentation accuracy and calibration across both datasets and architectures, while substantially reducing training time, inference latency, memory footprint, and storage requirements. Against efficient baselines - Snapshot Ensemble, MC Dropout, and Deep Deterministic Uncertainty - ST-LoRA consistently matches or outperforms them in image/pixel-level OoD detection, calibration stability under shift, and cross-seed variance, with far fewer parameters and lower compute. These results show LoRA-efficient ensemble adaptation is a highly effective, practical approach for uncertainty-aware agricultural vision systems.
Comments:Submitted to Computers and Electronics in Agriculture (Elsevier). Currently under review (first revision round)
Subjects:Computer Vision and Pattern Recognition (cs.CV)
Cite as:arXiv:2608.01530 [cs.CV]
(or arXiv:2608.01530v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.01530
arXiv-issued DOI via DataCite
Submission history
From: Mohamed Farag [view email]
[v1] Sun, 2 Aug 2026 22:49:28 UTC (22,114 KB)