Search papers, labs, and topics across Lattice.
AgentConductor introduces a reinforcement learning-optimized multi-agent system with an LLM-based orchestrator that dynamically generates interaction topologies for competition-level code generation. It uses a novel topological density function and difficulty interval partitioning to create task-adapted, density-aware layered DAG topologies. Experiments on code generation datasets show AgentConductor achieves state-of-the-art accuracy with significant reductions in topology density and token cost compared to existing methods.
Forget static agent communication graphs: AgentConductor uses RL to dynamically rewire agent interactions based on task difficulty, slashing token costs by up to 68% while boosting code generation accuracy.
Large language model(LLM)-driven multi-agent systems(MAS) coordinate specialized agents through predefined interaction topologies and have shown promise for complex tasks such as competition-level code generation. Recent studies demonstrate that carefully designed multi-agent workflows and communication graphs can significantly improve code generation performance by leveraging collaborative reasoning. However, existing methods neither adapt topology density to task difficulty nor iteratively refine the topology within an instance using execution feedback, which leads to redundant communication and performance bottlenecks. To address these issues, we propose AgentConductor: a reinforcement learning-optimized MAS with an LLM-based orchestrator agent as its core, which enables end-to-end feedback-driven dynamic generation of interaction topologies. For each query, AgentConductor infers agent roles and task difficulty, then constructs a task-adapted, density-aware layered directed acyclic graph (DAG) topology, underpinned by two key innovations. First, we design a novel topological density function that captures communication-aware mathematical characterizations of multi-agent interactions. Second, we adopt difficulty interval partitioning to avoid excessive pruning for precise topological density upper bound measurement per difficulty level and finer-grained control. Empirically, across three competition-level and two foundational code datasets, AgentConductor achieves state-of-the-art accuracy, outperforming the strongest baseline by up to 14.6% in pass@1 accuracy, 13% in density reduction, and 68% in token cost reduction.
