A Systematic Study on Pretraining Strategies for Low-Label Remote Sensing Image Semantic Segmentation.

Sensors (Basel) · 2026 · PMC12944012 · PMID 41755323

Europe PMC DOI

hedged sentence

—

closest p

—

boldest claim

The sentences

an interesting trendno p-value reported

The results also reveal an interesting trend: at lower annotation ratios, ViT-based models often underperform their ResNet-50 counterparts.

also in 1,024 other papers

Quoted from the open-access full text in Europe PMC under the licence the publisher applied. The sentence is reproduced exactly as published; the emphasis is ours.