A Traditional Approach to Symbolic Piano Continuation
This work addresses the MIREX 2025 Symbolic Music Generation challenge, offering an incremental improvement by applying a traditional method to a specific domain.
The authors tackled symbolic piano music continuation by using a simple next-token-prediction approach on tokenized MIDI data, arguing it outperforms large foundation models for constrained tasks, though no concrete performance numbers are provided in the abstract.
We present a traditional approach to symbolic piano music continuation for the MIREX 2025 Symbolic Music Generation challenge. While computational music generation has recently focused on developing large foundation models with sophisticated architectural modifications, we argue that simpler approaches remain more effective for constrained, single-instrument tasks. We thus return to a simple, unaugmented next-token-prediction objective on tokenized raw MIDI, aiming to outperform large foundation models by using better data and better fundamentals. We release model weights and code at https://github.com/christianazinn/mirex2025.