CSU learns per-class unreliability parameters to reduce class-dependent supervision bias from spurious, misassigned, or weak labels in audio tagging, with gains shown on AudioSet and a new ESC-FreeGen50 benchmark mixing real and generated audio.
Title resolution pending
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
-
Robust Audio Tagging under Class-wise Supervision Unreliability
CSU learns per-class unreliability parameters to reduce class-dependent supervision bias from spurious, misassigned, or weak labels in audio tagging, with gains shown on AudioSet and a new ESC-FreeGen50 benchmark mixing real and generated audio.