Of All StrIPEs: Investigating Structure-informed Positional Encoding for Efficient Music Generation
This work addresses the problem of computational efficiency and performance in music generation for AI researchers, though it is incremental as it builds on existing positional encoding techniques.
The paper tackled the challenge of efficient music generation by developing a unified kernel-based framework to analyze and compare different positional encoding methods, resulting in a novel method called RoPEPool that outperforms all existing methods in melody harmonization tasks.
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.