Roles with Rails: Contract-Preserving Role Evolution in Multi-Agent Structured Reasoning
For developers of LLM multi-agent systems, this work addresses the trade-off between adaptivity and structural integrity, offering a principled approach to role evolution.
The paper tackles the problem of adapting role pools in LLM multi-agent systems while preserving structural contracts. SERO achieves up to 15% improvement in task score on reasoning benchmarks across three LLM backbones.
Role-based LLM multi-agent systems need adaptive role pools, yet adapting such systems is not merely a matter of prompt optimization: roles often carry structural obligations, including capability coverage, message compatibility, validation, final-answer aggregation, and parser-compatible output protocols. Existing systems either fix the role inventory and lose adaptivity, or allow unconstrained generation to induce role drift, removing structurally necessary roles and breaking answer contracts. We formulate this as contract-preserving role evolution, requiring every committed edit to preserve five structural contracts (capability, communication, validation, aggregation, output protocol). We instantiate this formulation in SERO, a Self-Evolving Role Orchestration framework that evolves a typed role-card pool through credit-guided retrieval, a credit-ranked communication DAG with a protected terminal aggregator and conditional validator repair, and a contextual-bandit controller whose LLM-proposed edits are committed only when they preserve the contracts and improve task score. Experiments on real-world reasoning benchmarks across three LLM backbones confirm the value of contract-preserving role evolution.