الملخص
When does distribution alignment help a frozen foundation-model embedding generalize across acoustic domains? For cross-domain mosquito-species classification we report a strength-monotonic law: the stronger an encoder is on the target task, the more its unseen-domain generalization relies on a distribution-alignment (MMD) term, and the more it is harmed by domain-rebalanced sampling. Across four encoder families and a within-encoder HuBERT layer sweep (n=8), the rebalancing leg orders exactly with encoder strength (Spearman -1.000), while the MMD-benefit leg is monotonic within each stream and -0.857 pooled; fixing architecture and varying only representation strength flips the rebalancing effect from benefit to collapse. The law is actionable: a single MMD term is the sole lever on a strong encoder, so we reduce the field's default recipe to a frozen Perch 2.0 embedding, a lightweight probe, cross-entropy, one MMD, and input augmentation. The reduced recipe stays within seed noise of the full composite (BA_unseen 0.299+/-0.006 vs. 0.307+/-0.014). As boundary conditions of the same law, three community defaults (backbone fine-tuning, multi-modal fusion, and domain rebalancing) each hurt unseen-domain accuracy under a leave-domain protocol, shown with single-variable, multi-seed evidence. We present a mechanism and the recipe it explains, not a leaderboard entry.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Gong, Y., Zhou, R., Zeng, B., Ren, Q., & Hui, H. (2026). A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification. https://omanscience.com/ar/articles/a-strength-monotonic-law-for-domain-alignment-in-frozen-embedding-bioacoustic-classification
MLA 9
Gong, Yucheng, et al. "A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification." https://omanscience.com/ar/articles/a-strength-monotonic-law-for-domain-alignment-in-frozen-embedding-bioacoustic-classification.
شيكاغو (المؤلف–التاريخ)
Gong, Yucheng, Rui Zhou, Binbin Zeng, Qiang Ren, and Hongjin Hui. 2026. "A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification." https://omanscience.com/ar/articles/a-strength-monotonic-law-for-domain-alignment-in-frozen-embedding-bioacoustic-classification.
هارفارد
Gong, Y., Zhou, R., Zeng, B., Ren, Q. and Hui, H. (2026) 'A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification', Available at: https://omanscience.com/ar/articles/a-strength-monotonic-law-for-domain-alignment-in-frozen-embedding-bioacoustic-classification.
فانكوفر
Gong Y, Zhou R, Zeng B, Ren Q, Hui H. A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification. https://omanscience.com/ar/articles/a-strength-monotonic-law-for-domain-alignment-in-frozen-embedding-bioacoustic-classification
IEEE
Y. Gong, R. Zhou, B. Zeng, Q. Ren, and H. Hui, "A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification," https://omanscience.com/ar/articles/a-strength-monotonic-law-for-domain-alignment-in-frozen-embedding-bioacoustic-classification.