发表机构
Virginia Tech（弗吉尼亚理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出Evo2Team方法，通过选择、适应和确认源技能实现多智能体技能迁移，在28个迁移方向中20个达到正迁移标准，并显著降低部署成本，强调需从行动、任务及部署质量评估迁移效果。
AI 中文摘要
一个技能库可能帮助一个多智能体系统，却让另一个系统的行为保持不变。只有当目标智能体成功执行迁移的规则时，该规则才有帮助。我们研究了Count-Frequency和AgentsNet中的路由与通信技能，使用了4至32个智能体的团队以及GPT和Qwen模型阶梯。在16个设置中的14个里，源进化满足了联合质量、成本、模型层级和确认目标。随后，我们评估了Evo2Team，它为目标团队选择、适应并确认源技能，并与六个冻结选择器在28个迁移方向上进行了对比。Evo2Team的目标侧探索成本在每个方向上都低于进化新目标技能库的成本，即使重复使用的参考评估被计费一次。28个保留结果中有20个满足正迁移标准，包括三个节省的诊断测试。仅选择并不能解释这些结果：KNN和CORAL在两个AgentsNet方向上选择了不同的技能库，但产生了相同的记录执行。当Evo2Team改变执行时，收益可覆盖多个任务，例如在一个Count-Frequency方向上，相比KNN改善了28个任务中的32个。七个正迁移的AgentsNet结果节省了6.1%至14.6%的部署成本，同时仅在15个任务中的三至六个上使用迁移技能。在五个早期接受的方向中，使用迁移技能的所有22个任务记录通过了三个固定图确认，但在新图上的记录执行中有四个失败。在此比较中，图与模型响应同时变化。这些结果表明，技能迁移必须通过智能体采取的行动、这些行动所触及的任务以及最终部署的质量和成本来评估。
英文摘要
A skill bank that helps one multi-agent system may leave another's behavior unchanged. A transferred rule helps only when target agents act on it successfully. We study this path for routing and communication skills in Count-Frequency and AgentsNet, using teams of 4--32 agents and GPT and Qwen model ladders. Source evolution meets a joint quality, cost, model-tier, and confirmation goal in 14 of 16 settings. We then evaluate Evo2Team, which selects, adapts, and confirms source skills for the target team, alongside six frozen selectors across 28 transfer directions. Evo2Team's target-side exploration cost is below that of evolving a new target bank in every direction, even when reused reference evaluations are charged once. Twenty of 28 held-out outcomes meet the positive-transfer criterion, including three saved diagnostic tests. Selection alone does not explain these outcomes: KNN and CORAL choose different banks in two AgentsNet directions but produce identical recorded executions. When Evo2Team changes execution, gains can reach many tasks, as in a Count-Frequency direction that improves 28 of 32 tasks over KNN. Seven positive AgentsNet outcomes save 6.1--14.6\% in deployment cost while using transferred skills on only three to six of fifteen tasks. In five earlier accepted directions, all 22 task records using transferred skills pass three fixed-graph confirmations, but four fail in recorded executions on new graphs. Graphs and model responses change together in this comparison. These results show that skill transfer must be assessed through the actions agents take, the tasks those actions reach, and the quality and cost of the final deployment.
Comments26 pages, 11 figures
