Mask polarization restores bimodality in SE model predictions via Wasserstein distance at test time, delivering consistent gains across domain shifts and architectures.
Results Analysis Table 1 presents the results averaged across all target datasets for the AM architecture
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Test-Time Adaptation For Speech Enhancement Via Mask Polarization
Mask polarization restores bimodality in SE model predictions via Wasserstein distance at test time, delivering consistent gains across domain shifts and architectures.