2026
Unmute the Patch Tokens: Rethinking Probing in Multi-Label Audio Classification
ICLR 2026poster
Although probing frozen models has become a standard evaluation paradigm, self-supervised learning in audio defaults to fine-tuning when pursuing state-of-the-art on AudioSet. A key reason is that global pooling creates an information bottleneck causing linear probes to misrepresent the embedding qu…