[Submitted on 31 Oct 2026 (v1), last revised 3 Nov 2026 (this version, v2)]
Abstract:Composers develop music from an original idea, controlling how similar each new section is to it, from near repetition to new material. Game music needs this often, as its material repeats over long play. Music models, however, rarely learn to control similarity from real pieces. We further train a pretrained MIDI model on section pairs from Super Mario pieces. A label states each pair’s similarity, and a context vector summarises the given section, or context. For 157 test contexts, the model generates six outputs per label. Twelve measures score the outputs; one main measure each covers melody, pitch content, rhythm and note length. On average, outputs requested as similar are more similar to their context than those requested as dissimilar, and not to other pieces. All twelve measures favour them; each main measure stays significant after correction for multiple tests. On melody, the gap comes mainly from more frequent restatement of the context’s melody. Against an otherwise identical model without a label, only the request for high similarity makes outputs more similar. In a listening test, ratings followed melodic overlap with the context most closely of all measures.
Submission history
From: Shangxuan Luo [view email]
[v1] Sat, 31 Oct 2026 15:40:27 UTC (282 KB)
[v2] Tue, 3 Nov 2026 12:00:30 UTC (282 KB)
