PML-MA restores feature-label consistency in partial multi-label learning by generating pseudo-labels via low-rank orthogonal decomposition, aligning modalities globally and locally, and using multi-peak prototypes for refinement.
Partial multi-label learning via multi- subspace representation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Feature-Label Modal Alignment for Robust Partial Multi-Label Learning
PML-MA restores feature-label consistency in partial multi-label learning by generating pseudo-labels via low-rank orthogonal decomposition, aligning modalities globally and locally, and using multi-peak prototypes for refinement.