自動化代理組合：基於背包問題的代理組件選擇方法

Automated Composition of Agents: A Knapsack Approach for Agentic Component Selection

October 18, 2025

作者: Michelle Yuan, Khushbu Pahwa, Shuaichen Chang, Mustafa Kaba, Jiarong Jiang, Xiaofei Ma, Yi Zhang, Monica Sunkara

cs.AI

摘要

設計有效的代理系統需要在動態且不確定的環境中無縫地組合和整合代理、工具和模型。現有方法大多依賴於靜態的語義檢索方法來發現工具或代理。然而，由於能力描述的不完整性和檢索方法的局限性，現有組件的有效重用和組合仍然具有挑戰性。組件選擇的問題在於決策並未基於能力、成本和實時效用。為解決這些挑戰，我們引入了一個受背包問題啟發的結構化、自動化框架，用於代理系統的組合。該框架使組合代理能夠系統地識別、選擇並組裝一組最佳的代理組件，同時考慮性能、預算限制和兼容性。通過動態測試候選組件並實時建模其效用，我們的方法簡化了代理系統的組裝，並促進了資源的可擴展重用。使用Claude 3.5 Sonnet在五個基準數據集上的實證評估表明，基於在線背包的組合器始終位於帕累托前沿，與基線相比，在顯著降低組件成本的情況下實現了更高的成功率。在單代理設置中，在線背包組合器的成功率相比檢索基線提高了高達31.6%。在多代理系統中，當從包含100多個代理的代理庫中選擇代理時，在線背包組合器的成功率從37%提升至87%。顯著的性能差距證實了我們的方法在不同領域和預算限制下的強大適應性。

English

Designing effective agentic systems requires the seamless composition and integration of agents, tools, and models within dynamic and uncertain environments. Most existing methods rely on static, semantic retrieval approaches for tool or agent discovery. However, effective reuse and composition of existing components remain challenging due to incomplete capability descriptions and the limitations of retrieval methods. Component selection suffers because the decisions are not based on capability, cost, and real-time utility. To address these challenges, we introduce a structured, automated framework for agentic system composition that is inspired by the knapsack problem. Our framework enables a composer agent to systematically identify, select, and assemble an optimal set of agentic components by jointly considering performance, budget constraints, and compatibility. By dynamically testing candidate components and modeling their utility in real-time, our approach streamlines the assembly of agentic systems and facilitates scalable reuse of resources. Empirical evaluation with Claude 3.5 Sonnet across five benchmarking datasets shows that our online-knapsack-based composer consistently lies on the Pareto frontier, achieving higher success rates at significantly lower component costs compared to our baselines. In the single-agent setup, the online knapsack composer shows a success rate improvement of up to 31.6% in comparison to the retrieval baselines. In multi-agent systems, the online knapsack composer increases success rate from 37% to 87% when agents are selected from an agent inventory of 100+ agents. The substantial performance gap confirms the robust adaptability of our method across diverse domains and budget constraints.