Abstract
Large language models (LLMs) enable autonomous agents for reasoning, planning, and tool use. Recent systems increasingly organize these agents as graphs of specialized, interconnected nodes. Although graph-based orchestration supports flexible decomposition and coordination, it creates a key challenge: attention allocation. As workflows grow, existing approaches often execute graph components uniformly, wasting resources on irrelevant or low-impact tasks.
We introduce Attention Orchestration, a paradigm that extends Transformer-style attention from token representations to workflow-level agent coordination. Our framework, Adaptive Goal-aware Attention Orchestration (AGAO), dynamically estimates agent importance based on user objectives, graph dependencies, and computational constraints. AGAO combines three components:
- Goal-aware attention: measuring semantic relevance between user goals and agent capabilities;
- Topology-aware attention: modeling structural dependencies in agent graphs;
- Resource-aware attention: allocating budgets and execution priorities across heterogeneous agents.
Together, these mechanisms transform static agent graphs into adaptive systems that focus computation on goal-critical reasoning paths. Experiments across diverse multi-agent workloads show that AGAO improves task effectiveness while reducing unnecessary computation, latency, and token consumption compared with existing graph-based execution strategies. Our work establishes Attention Engineering as a direction for scalable, intelligent multi-agent systems.
Code: AGAO GitHub
Blogger's Review: This research significantly enhances the resource efficiency of multi-agent systems by introducing adaptive goal-aware attention orchestration, showcasing the potential for collaboration among agents. As the application of multi-agent systems expands, attention engineering will become one of the key technologies worth exploring further.