Soft-label distillation from same-architecture teacher ensembles improves respiratory sound classification and sets a new ICBHI score of 64.39, though gains are partly due to test-set-based selection of settings.
Previous works primarily focused on CNN architectures [1–8], such as ResNet [9], EfficientNet [10], and CNN6 [11]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Improving Respiratory Sound Classification with Architecture-Agnostic Knowledge Distillation from Ensembles
Soft-label distillation from same-architecture teacher ensembles improves respiratory sound classification and sets a new ICBHI score of 64.39, though gains are partly due to test-set-based selection of settings.