|国家预印本平台
首页|Of All StrIPEs: Investigating Structure-informed Positional Encoding for Efficient Music Generation

Of All StrIPEs: Investigating Structure-informed Positional Encoding for Efficient Music Generation

Of All StrIPEs: Investigating Structure-informed Positional Encoding for Efficient Music Generation

来源:Arxiv_logoArxiv
英文摘要

While music remains a challenging domain for generative models like Transformers, a two-pronged approach has recently proved successful: inserting musically-relevant structural information into the positional encoding (PE) module and using kernel approximation techniques based on Random Fourier Features (RFF) to lower the computational cost from quadratic to linear. Yet, it is not clear how such RFF-based efficient PEs compare with those based on rotation matrices, such as Rotary Positional Encoding (RoPE). In this paper, we present a unified framework based on kernel methods to analyze both families of efficient PEs. We use this framework to develop a novel PE method called RoPEPool, capable of extracting causal relationships from temporal sequences. Using RFF-based PEs and rotation-based PEs, we demonstrate how seemingly disparate PEs can be jointly studied by considering the content-context interactions they induce. For empirical validation, we use a symbolic music generation task, namely, melody harmonization. We show that RoPEPool, combined with highly-informative structural priors, outperforms all methods.

Manvi Agarwal、Changhong Wang、Gael Richard

LTCILTCIS2A, IDS

计算技术、计算机技术

Manvi Agarwal,Changhong Wang,Gael Richard.Of All StrIPEs: Investigating Structure-informed Positional Encoding for Efficient Music Generation[EB/OL].(2025-04-07)[2025-05-28].https://arxiv.org/abs/2504.05364.点此复制

评论