MusicSwarm: Biologically Inspired Intelligence for Music Composition
This provides a compute- and data-efficient method for long-horizon creative tasks like music composition, with potential applications in collaborative writing and design, though it is incremental in applying swarm intelligence to a new domain.
The paper tackled the problem of generating coherent, long-form music composition by using a decentralized swarm of identical, frozen foundation models that coordinate without weight updates, resulting in superior quality, greater diversity, and structural variety compared to a centralized system.
We show that coherent, long-form musical composition can emerge from a decentralized swarm of identical, frozen foundation models that coordinate via stigmergic, peer-to-peer signals, without any weight updates. We compare a centralized multi-agent system with a global critic to a fully decentralized swarm in which bar-wise agents sense and deposit harmonic, rhythmic, and structural cues, adapt short-term memory, and reach consensus. Across symbolic, audio, and graph-theoretic analyses, the swarm yields superior quality while delivering greater diversity and structural variety and leads across creativity metrics. The dynamics contract toward a stable configuration of complementary roles, and self-similarity networks reveal a small-world architecture with efficient long-range connectivity and specialized bridging motifs, clarifying how local novelties consolidate into global musical form. By shifting specialization from parameter updates to interaction rules, shared memory, and dynamic consensus, MusicSwarm provides a compute- and data-efficient route to long-horizon creative structure that is immediately transferable beyond music to collaborative writing, design, and scientific discovery.