Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chang, Chih-Heng, Ho, Keng-Seng, Tsai, Chih-Yu, Chen, Kuan-Lin, Yang, Yi-Hsuan, Ding, Jian-Jiun
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.31053
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913172359741440
author Chang, Chih-Heng
Ho, Keng-Seng
Tsai, Chih-Yu
Chen, Kuan-Lin
Yang, Yi-Hsuan
Ding, Jian-Jiun
author_facet Chang, Chih-Heng
Ho, Keng-Seng
Tsai, Chih-Yu
Chen, Kuan-Lin
Yang, Yi-Hsuan
Ding, Jian-Jiun
contents Controllable music editing is to modify high-level attributes while strictly preserving rhythmic and melodic structures. However, this task is challenged by a semantic-structural entanglement: steering methods often degrade structure to achieve editing performance, while structural adaptors suppress semantic responsiveness. We propose AnchorSteer, a framework that disentangles this tension by coupling structural anchoring with self-discovered semantic steering. The proposed approach probes internal representations to extract interpretable, label-free concept vectors via a self-supervised reconstruction objective, isolating attributes without curated data. During editing, these portable, plug-and-play concept vectors are injected into diffusion hidden manifolds while a structural adaptor enforces consistency. Variants for unconditioned and conditioned injections are provided to balance robustness and semantic strength. Experiments on ZoME-Bench and subjective tests show that the proposed framework outperforms both steering-only and anchoring-only baselines, enabling significant semantic transformations with high-fidelity structural preservation.
format Preprint
id arxiv_https___arxiv_org_abs_2605_31053
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle AnchorSteer: Self-Discovered Concept Injection for Structure-Preserving Music Editing
Chang, Chih-Heng
Ho, Keng-Seng
Tsai, Chih-Yu
Chen, Kuan-Lin
Yang, Yi-Hsuan
Ding, Jian-Jiun
Sound
Artificial Intelligence
Controllable music editing is to modify high-level attributes while strictly preserving rhythmic and melodic structures. However, this task is challenged by a semantic-structural entanglement: steering methods often degrade structure to achieve editing performance, while structural adaptors suppress semantic responsiveness. We propose AnchorSteer, a framework that disentangles this tension by coupling structural anchoring with self-discovered semantic steering. The proposed approach probes internal representations to extract interpretable, label-free concept vectors via a self-supervised reconstruction objective, isolating attributes without curated data. During editing, these portable, plug-and-play concept vectors are injected into diffusion hidden manifolds while a structural adaptor enforces consistency. Variants for unconditioned and conditioned injections are provided to balance robustness and semantic strength. Experiments on ZoME-Bench and subjective tests show that the proposed framework outperforms both steering-only and anchoring-only baselines, enabling significant semantic transformations with high-fidelity structural preservation.
title AnchorSteer: Self-Discovered Concept Injection for Structure-Preserving Music Editing
topic Sound
Artificial Intelligence
url https://arxiv.org/abs/2605.31053