Ecologically-Constrained Task Arithmetic for Multi-Taxa Bioacoustic Classifiers Without Shared Data

TL;DR

This paper demonstrates that bioacoustic classifiers can be composed from independently fine-tuned encoders using task vector arithmetic, enabling collaborative, privacy-preserving biodiversity monitoring across taxa and regions.

Contribution

It introduces a method for composing multi-taxa classifiers via task vectors without sharing data, leveraging near-orthogonal task vectors and spectral distribution analysis.

Findings

Task vectors are near-orthogonal with cosine similarity 0.01-0.09.

Simple averaging of task vectors is optimal for composition.

The approach enables zero-shot transfer and preserves data privacy.

Abstract

Training data for bioacoustics is scattered across taxa, regions, and institutions. Centralizing it all is often infeasible. We show that independently fine-tuned BEATs encoders can be composed into a unified 661-species classifier via task vector arithmetic without sharing data. We find that bioacoustic task vectors are near-orthogonal (cosine 0.01-0.09). Their separation aligns closely with spectral distribution distance, a gradient consistent with the acoustic niche hypothesis. This geometry makes simple averaging optimal while sign-conflict methods reduce accuracy by one to six percentage points. Composition also creates an asymmetric gap: species-rich groups lose accuracy relative to joint training while underrepresented taxa gain, a redistribution useful for equitable biodiversity monitoring. We verify linear mode connectivity across all taxonomic pairs, demonstrate zero-shot transfer to new regions, and identify domain negation as a boundary condition where composition fails. These results enable a collaborative paradigm for bioacoustics where institutions share only task vectors to assemble multi-taxa classifiers, preserving data privacy.