[论文解读] SC-MAS: Constructing Cost-Efficient Multi-Agent Systems with Edge-Level Heterogeneous Collaboration

Di Zhao, Longhui Ma+3 more|arXiv (Cornell University)|Jan 14, 2026

Advanced Graph Neural Networks被引用 0

一句话总结

SC-MAS 通过选择任务特定的代理角色、边缘级协作策略和 LLM 骨干，构建可执行的异质多代理系统，在成本更低的同时达到比最先进方法更高的准确性。

ABSTRACT

Large Language Model (LLM)-based Multi-Agent Systems (MAS) enhance complex problem solving through multi-agent collaboration, but often incur substantially higher costs than single-agent systems. Recent MAS routing methods aim to balance performance and overhead by dynamically selecting agent roles and language models. However, these approaches typically rely on a homogeneous collaboration mode, where all agents follow the same interaction pattern, limiting collaboration flexibility across different roles. Motivated by Social Capital Theory, which emphasizes that different roles benefit from distinct forms of collaboration, we propose SC-MAS, a framework for constructing heterogeneous and cost-efficient multi-agent systems. SC-MAS models MAS as directed graphs, where edges explicitly represent pairwise collaboration strategies, allowing different agent pairs to interact through tailored communication patterns. Given an input query, a unified controller progressively constructs an executable MAS by selecting task-relevant agent roles, assigning edge-level collaboration strategies, and allocating appropriate LLM backbones to individual agents. Experiments on multiple benchmarks demonstrate the effectiveness of SC-MAS. In particular, SC-MAS improves accuracy by 3.35% on MMLU while reducing inference cost by 15.38%, and achieves a 3.53% accuracy gain with a 12.13% cost reduction on MBPP. These results validate the feasibility of SC-MAS and highlight the effectiveness of heterogeneous collaboration in multi-agent systems.

研究动机与目标

• 为 MAS 设计提供受社交资本启发的异质协作框架的动机。
• 联合优化代理选择、边缘级协作策略与 LLM 分配，以在准确性与成本之间取得平衡。
• 证明边缘级异质性在基准测试中提升性能并降低推理成本。

提出的方法

• 将 MAS 建模为有向无环图，边编码成对协作策略。
• 采用三阶段构建：节点选择器用于选择与任务相关的代理，边缘优化器在满足 DAG 约束的情况下分配代理之间的策略和自环策略，LLM 路由器通过图神经网络分配 LLM 骨干。
• 通过策略梯度优化策略，最大化任务效用减去成本惩罚，支持基于查询的图构建。
• 使用变分潜变量模型进行节点选择与概率化的边策略分配，以捕获协作上下文。
• 利用 GNN 传播结构上下文，并基于节点嵌入与查询特征计算 LLM 分配概率。

实验结果

研究问题

• RQ1 如何用边缘级协作策略表示 MAS，以捕捉异质交互？
• RQ2 联合优化代理选择、边缘策略和 LLM 分配是否能带来比同质 MAS 设计更好的准确性-成本权衡？
• RQ3 基于社交资本启发、基于 DAG 的 MAS 构建是否能在基准测试中提升效率与性能？

主要发现

方法	LLM 骨干	MAS 路由	MMLU	GSM8K	MATH	HumanEval	MBPP
Vanilla	gpt-4o-mini		77.81	93.17	66.09	85.71	72.20
claude-3.5-haiku			67.97	92.16	65.89	86.33	73.40
gemini-1.5-flash			80.04	92.67	74.39	82.61	73.00
llama-3.1-70b			79.08	92.68	60.31	80.75	68.20
GPTSwarm Zhuge et al. (2024)	gpt-4o-mini		82.80	94.66	68.85	86.28	75.40
gemini-1.5-flash			83.22	93.98	73.35	82.36	74.80
AgentPrune Zhang et al. (2025b)	gpt-4o-mini		83.02	94.89	68.45	86.80	75.40
gemini-1.5-flash			83.10	93.88	73.54	82.55	75.80
AFlow Zhang et al. (2025c)	gpt-4o-mini		83.10	92.30	73.35	90.06	82.20
gemini-1.5-flash			82.35	94.91	72.70	85.69	76.00
PromptLLM Feng et al. (2025)	LLM Pool		78.43	93.92	73.03	86.33	73.60
RouteLLM Ong et al. (2025)	LLM Pool		81.04	93.42	71.29	83.85	72.60
FrugalGPT Chen et al. (2024b)	LLM Pool		76.24	90.76	67.05	87.31	74.40
RouterDC Chen et al. (2024c)	LLM Pool		82.01	93.68	73.46	87.75	75.20
MasRouter Yue et al. (2025)	LLM Pool		84.25	95.45	75.42	90.62	84.00
SC-MAS (Ours)	LLM Pool		87.60	96.09	76.75	92.37	87.53

• SC-MAS 在多个基准上实现比基线更高的准确性和更低的成本（例如，MMLU 提升 3.35%，MBPP 成本降低 15.38%）。
• 在 MBPP 上，SC-MAS 提供 3.53% 的准确性提升，推理成本降低 12.13%。
• 消融实验表明移除边缘或 LLM 组件会降低性能并提高成本；边缘建模和 LLM 多样性至关重要。
• SC-MAS 能有效筛选无用的代理，对嘈杂的协作者给出较低的选择概率。
• 归纳分析显示 SC-MAS 能推广至未见过的 LLM，当引入更强模型时性能得到提升。