arXiv：cs.LG· Songtao Wei, Yi Li, Zhichun Guo, Bingzhe Li·· 3 小时前AI 评分38
Inherit-MAS：通过工作流与执行继承实现多智能体系统的测试时进化
Inherit-MAS: Test-Time Evolution of Multi-Agent Systems through Workflow and Execution Inheritance
AI 导读
Inherit-MAS 提出工作流继承与执行继承两种机制，让多智能体系统在测试时进化：前者从最新候选工作流出发，剔除无用节点并做经验证的编辑；后者在完整请求与执行上下文匹配时复用已存结果，避免重复模型与工具调用。
正文
Abstract:Multi-agent systems (MAS) built from large language models coordinate specialized agents to tackle complex tasks, but effective workflows are difficult to design in advance. Test-time evolution refines workflows using execution feedback, yet broad revisions can disturb useful components, while re-executing unchanged requests can incur redundant computation. Inspired by the interplay of inheritance and selection in biological evolution, we introduce Inherit-MAS, which makes inheritance explicit at the workflow and execution levels. A meta-model first synthesizes a workflow of worker agents with declared roles, communication inputs, and tool permissions, and a separately prompted judge scores each executed candidate and diagnoses its deficiencies. In ordinary refinement rounds, \emph{workflow inheritance} starts from the latest completed candidate, may discard removable nodes judged unhelpful, and applies a validated edit to address the diagnosed deficiency. When the new candidate executes, \emph{execution inheritance} inherits eligible stored results only if the complete resolved request and execution context match, avoiding redundant model and tool calls. With GPT-4o-mini workers, Inherit-MAS achieves 55.4\% completion on WorkBench and 49.7\% joint F1 on HotpotQA FullWiki, outperforming EvoAgent, EvoMAS, and TacoMAS. With Qwen3-32B workers, it also exceeds these evolving-MAS baselines on both benchmarks. Compared with rerunning the same controller with execution inheritance disabled, execution inheritance reduces worker-token usage by 29.1\% on WorkBench and 34.6\% on HotpotQA, and total token usage by 5.3\% and 18.1\%.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.02396 [cs.LG] |
| (or arXiv:2610.02396v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02396 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Songtao Wei [view email]
[v1]
Thu, 1 Oct 2026 19:23:32 UTC (412 KB)
来源：arXiv：cs.LG · arxiv.org