A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification

Yucheng Gong; Rui Zhou; Binbin Zeng; Qiang Ren; Hongjin Hui

eess.ASOct 7, 2026

A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification

Authors: Yucheng Gong, Rui Zhou, Binbin Zeng, Qiang Ren, Hongjin Hui

Organizations: Department of Software Engineering, School of Software, Chongqing Finance and Economics College Chongqing Key Laboratory of Data Mining and Integrated Application for Ecological Environment

Abstract

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.

Figures & tables

Method	$\mathrm{BA}_{\mathrm{unseen}}$	Note
Frozen Perch + RP + full composite (seed-ensemble)	0.313	champion base $0.318$ ; its per-seed $\sigma$ unavailable
same, per-seed mean	$0.307\pm 0.014$
Frozen Perch + RP + MMD-only, per-seed mean	$0.299\pm 0.006$	reduced recipe, within noise of full
same, seed-ensemble	$0.308$
— no augmentation (loss only)	$0.274\pm 0.003$
— cross-entropy only (aug, no alignment)	$0.235\pm 0.002$	architecture alone is not the win

Table 1 : Main leave-domain ablation.

\mathrm{BA}_{\mathrm{unseen}}

as per-seed

\mu\pm\sigma

unless marked as a seed-ensemble. The deployed champion system (

0.362

) adds a three-feature agreement gate that is orthogonal to this study; we compare only to its frozen single-modal base.

Arm	$\mathrm{BA}_{\mathrm{unseen}}$	$\Delta$ vs full	Verdict
full	$0.307\pm 0.014$	—	—
drop-CORAL	$0.308\pm 0.009$	$+0.001$	removable
drop-SupCon	$0.313\pm 0.006$	$+0.006$	removable
drop-genus	$0.305\pm 0.010$	$-0.003$	removable (within noise)
drop-domain	$0.307\pm 0.012$	$-0.001$	removable (within noise)
drop-MMD	$\mathbf{0.209\pm 0.004}$	$\mathbf{-0.098}$	the only lever

Table 2 : Leave-one-loss ablation on frozen Perch (

\mathrm{BA}_{\mathrm{unseen}}

\mu\pm\sigma

). Only MMD matters.

Config	Perch (CNN-mel)	BEATs (transf-mel)	BirdAVES (wav-SSL)	HuBERT (speech-SSL)
full	$0.307\pm 0.014$	$0.259\pm 0.009$	$0.188\pm 0.008$	$0.165\pm 0.014$
drop-MMD	$0.209\pm 0.004$	$0.189\pm 0.002$	$0.140\pm 0.001$	$0.150\pm 0.003$
MMD contribution	$\mathbf{-0.098}$	$\mathbf{-0.070}$	$\mathbf{-0.048}$	$\mathbf{-0.015}$
rebalance $\Delta$ (vs random)	$\mathbf{-0.072}$	$\mathbf{-0.056}$	$\mathbf{+0.015}$	$\mathbf{+0.034}$

Table 3 : Cross-encoder ablation (30 epochs, leave-domain,

\mu\pm\sigma

). Perch/BEATs/BirdAVES use seeds 42/0/1; HuBERT uses 42/43/44 to match the layer sweep. The rebalancing row orders exactly in encoder strength; the MMD row is monotonic within this stream.

Figure 1 : The strength-monotonic law. Domain-rebalancing harm orders exactly with encoder task strength (Spearman

-1.000

n{=}8

); MMD benefit grows with strength, monotonic within each stream and

-0.857

pooled.

HuBERT layer	$\mathrm{BA}_{\mathrm{unseen}}$ (strength)	MMD contribution	rebalance $\Delta$ (vs random)
L12 (last)	$0.165\pm 0.014$	$-0.015$	$+0.034$ gain
L9	$0.201\pm 0.016$	$-0.043$	$-0.014$ neutral
L8	$0.196\pm 0.009$	$-0.032$	$-0.005$ neutral
L7	$0.198\pm 0.008$	$-0.036$	$-0.007$ neutral
L6	$\mathbf{0.232\pm 0.008}$	$-0.058$	$\mathbf{-0.051}$ collapse

Table 4 : HuBERT within-layer sweep (architecture fixed, strength varied). The rebalancing effect flips sign from the weakest to the strongest layer.

Figure 2 : Within-encoder evidence. Holding HuBERT’s architecture fixed and varying only layer strength, the rebalancing effect flips from benefit (weak L12) to collapse (strong L6).

	Perch	BEATs	BirdAVES	HuBERT L12	L6	L7	L8	L9
domain separability	$0.874$	$0.917$	$0.885$	$0.778$	$0.844$	$0.812$	$0.815$	$0.791$

Table 5 : Domain-separability probe (linear five-way balanced accuracy; chance

0.20

). Separability is uniformly high and does not track collapse.

Explore similar work

Is Domain Adaptation Always Helpful? A Frozen-Backbone Study of Cross-Domain Sentiment Transfer
Jul 7, 2026Phat Tran, Artin Lahni, Pranav Kulkarni +1Domain-Adversarial Learning Domain Adaptation
Oregon State University
Scoring Backends Matter More Than Pooling: A Systematic Study of Training-Free Anomalous Sound Detection under Domain Shift
Jun 17, 2026Jingwen Zhou, Mingzhe WangDistribution Shift Unsupervised Anomaly Detection
Xidian University, Xi’an, China
Phylogenetic signal in marine mammal and bird vocalizations captured by audio foundation models: the limited benefit of domain-specific pretraining
Jul 24, 2026Víctor Rincón YepesAudio Representation Learning Bioacoustics
Independent Researcher

Jul 7, 2026, cs.CLJ/K move · Enter open · S save

Is Domain Adaptation Always Helpful? A Frozen-Backbone Study of Cross-Domain Sentiment Transfer

Phat Tran, Artin Lahni, Pranav Kulkarni, Yaolun Zhang

Oregon State University

Domain-Adversarial Learning Domain Adaptation Sentiment Analysis

Open paper AskPDF arXiv

Sentiment analysis with frozen pre-trained language model (PLM) backbones has become a common paradigm, yet the practical benefit of explicit domain adaptation remains unclear, particularly when backbones encode varying degrees of target-domain knowledge. We present a preliminary case study evaluating a controlled family of frozen embedding backbones (Qwen3-Embedding 0.6B, 4B, 8B), alongside RoBERTa-base and FinBERT. We train a lightweight MLP adapter on consumer reviews using Domain-Adversarial Neural Networks (DANN), Maximum Mean Discrepancy (MMD), and Supervised Contrastive Learning (SCL), and evaluate transfer to movie reviews (SST-2) and a heavily restricted subset of financial news (Financial PhraseBank). Within this constrained sample, we observe two distinct transfer patterns. On SST-2, domain adaptation provides negligible gain regardless of scale. On the financial subset, explicit domain adaptation appears to recover substantial performance for small general-purpose backbones. Notably, we find that adversarial alignment (DANN) is associated with degraded performance for domain-specialized backbones like FinBERT, consistent with erosion of pre-existing domain-specific structure, whereas supervised contrastive loss appears to preserve it. These preliminary findings suggest that the efficacy of explicit domain adaptation is highly contingent on whether the frozen backbone already possesses target-domain coverage.

Abstract

Explore similar work

Is Domain Adaptation Always Helpful? A Frozen-Backbone Study of Cross-Domain Sentiment Transfer

Scoring Backends Matter More Than Pooling: A Systematic Study of Training-Free Anomalous Sound Detection under Domain Shift

Phylogenetic signal in marine mammal and bird vocalizations captured by audio foundation models: the limited benefit of domain-specific pretraining