Researchers have developed Uno-Orchestra, a novel orchestration policy for large language model (LLM) multi-agent systems. This system learns to selectively decompose tasks and assign them to appropriate model-primitive pairs, optimizing decisions for task decomposition depth, worker choice, and inference budget simultaneously. In evaluations across 13 benchmarks, Uno-Orchestra outperformed 22 baselines, achieving a 16% higher macro pass@1 rate while reducing per-query costs by an order of magnitude. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Introduces a more efficient and accurate method for orchestrating LLM agents, potentially lowering costs for complex AI tasks.
RANK_REASON This is a research paper detailing a new method for LLM multi-agent systems. [lever_c_demoted from research: ic=1 ai=1.0]