MIFOMO adapts a remote sensing foundation model with coalescent projection, mixup domain adaptation, and label smoothing to outperform prior methods by up to 14% in cross-domain few-shot HSI classification.
Cross-domain few-shot learning with coalescent projections and latent space reservation
2 Pith papers cite this work. Polarity classification is still indexing.
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
CVLC fuses calibrated vision prototypes with LLM-generated language prototypes and applies dual coalescent projection plus latent space reservation to enable few-shot adaptation across sequential domains, reporting up to 16% gains over prior methods.
citing papers explorer
-
Cross-Domain Few-Shot Learning for Hyperspectral Image Classification Based on Mixup Foundation Model
MIFOMO adapts a remote sensing foundation model with coalescent projection, mixup domain adaptation, and label smoothing to outperform prior methods by up to 14% in cross-domain few-shot HSI classification.
-
Few-Shot Domain Incremental Learning via Continual Vision-Language Consolidation
CVLC fuses calibrated vision prototypes with LLM-generated language prototypes and applies dual coalescent projection plus latent space reservation to enable few-shot adaptation across sequential domains, reporting up to 16% gains over prior methods.