No AI summary available for this article.
Why It Matters
Few-shot audio classifiers may rely on foreground-background co-occurrences and fail when those correlations shift.
Provenance
Discovered via ArXiv and published by ArXiv.
Key Claims
Original description
Few-shot audio classifiers may rely on foreground-background co-occurrences and fail when those correlations shift. On SpurAudio, the resulting representation shift is concentrated and class dependent: for ResNet12, the top 10 percent of channels explain 82.80 percent of the null-corrected shift contribution. We propose SAMPLESELECT, which predicts a fixed-budget feature mask independently for each input while keeping the encoder and source classifier frozen. Training uses differentiable Gumbel Top-k selection with foreground classification and cross-background contrastive losses; inference us...
Discovered via ArXiv
Research papers and preprints from arXiv.
Publisher: arxiv.org
ID: http://arxiv.org/abs/2609.17076v1 · Indexed about 1 hour ago