MAFIG: Multi-agent Driven Formal Instruction Generation Framework

📄 arXiv: 2604.10989v1 📥 PDF

作者: Shixing Zhao, Zheng Si, Pengpeng Ouyang, Zhengqing Hu, Wanqi Zhu, Dong Chen, Yibo Guo, Mingliang Xu

分类: cs.AI

发布日期: 2026-04-13

💡 一句话要点

提出MAFIG框架以解决调度系统应急处理问题

🎯 匹配领域: 支柱二：RL算法与架构 (RL & Architecture) 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 调度系统 应急处理 多代理系统 正式指令生成 局部蒸馏 云大语言模型 系统稳定性 决策支持

📋 核心要点

  1. 现有调度系统在应急情况下常常依赖于预定义规则，难以适应复杂多变的实际场景。
  2. MAFIG框架通过局部功能模块的决策约束和快速生成正式指令来应对紧急情况。
  3. 实验结果显示，MAFIG在多个调度数据集上成功率高达98.49%，处理时间显著降低。

📝 摘要（中文）

调度系统中的紧急情况常常导致局部功能失效，影响系统稳定性。现有方法主要依赖于鲁棒调度或反应式调度，难以应对多样化和不可预测的紧急情况。为此，本文提出了多代理驱动的正式指令生成框架（MAFIG），通过生成正式指令快速修复调度逻辑。MAFIG包含感知代理和紧急决策代理，减少了系统上下文对决策的影响。同时，采用基于跨度的损失驱动局部蒸馏机制，将强大的云大语言模型的决策能力转移至轻量级本地模型，降低推理延迟。实验结果表明，MAFIG在多个调度数据集上取得了高达98.49%的成功率，显著提升了调度系统的鲁棒性和适应性。

🔬 方法详解

问题定义：本文旨在解决调度系统在紧急情况下的应急处理问题，现有方法因依赖预定义规则而难以适应多样化的紧急情况，导致系统稳定性受损。

核心思路：MAFIG框架通过限制决策范围到受影响的局部功能模块，快速生成正式指令以修复调度逻辑，从而提高应急处理的效率和准确性。

技术框架：MAFIG包含两个主要模块：感知代理用于实时监测系统状态，紧急决策代理负责生成应急指令。整体流程是先感知紧急情况，再通过决策代理生成相应的调度指令。

关键创新：引入了基于跨度的损失驱动局部蒸馏机制（SFL），将云大语言模型的决策能力转移到轻量级本地模型，显著降低推理延迟，同时保持决策效果。

关键设计：在模型设计中，采用了针对特定任务的损失函数和蒸馏策略，以确保轻量级模型在处理紧急情况时的高效性和准确性。具体参数设置和网络结构细节在实验部分进行了详细描述。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示，MAFIG在港口、仓储和甲板调度数据集上的成功率分别达到98.49%、94.97%和97.50%，平均处理时间分别为0.33秒、0.23秒和0.19秒，显著优于传统调度方法，展示了其在应急处理中的有效性。

🎯 应用场景

MAFIG框架具有广泛的应用潜力，尤其适用于港口、仓储和甲板调度等复杂调度系统。其快速应急处理能力能够显著提升系统的稳定性和适应性，未来可扩展至更多领域，如智能交通和供应链管理等。

📄 摘要（原文）

Emergency situations in scheduling systems often trigger local functional failures that undermine system stability and even cause system collapse. Existing methods primarily rely on robust scheduling or reactive scheduling, handling emergencies through predefined rules or rescheduling strategies. However, the diversity and unpredictability of real-world emergencies make them difficult to anticipate, which limits the adaptability of these methods in complex scenarios. Recent studies have shown that Large Language Models (LLMs) possess strong potential for complex scheduling tasks because of their extensive prior knowledge and strong reasoning capabilities. Nevertheless, the high inference latency of LLMs and the lengthy contextual information of scheduling systems significantly hinder their application for emergency handling. To mitigate these issues, we propose the Multi-agent Driven Formal Instruction Generation Framework (MAFIG). The framework constrains the decision scope to local functional modules affected by emergency situations and repairs scheduling logic rapidly by generating formal instructions. MAFIG contains a Perception Agent and an Emergency Decision Agent, which mitigates the adverse impact of lengthy system contexts on emergency decision-making. We further introduce span-focused loss-driven local distillation mechanism (SFL) to transfer the decision-making capability of powerful Cloud Large Language Models (C-LLMs) to lightweight local models, reducing inference latency while preserving decision-making effectiveness. Experiments in the Port, Warehousing, and Deck scheduling datasets show success rates of 98.49\%, 94.97\%, and 97.50\%, with average processing times of 0.33 s, 0.23 s, and 0.19 s. These results demonstrate that MAFIG effectively mitigates the impact of emergencies and improves the robustness and adaptability of scheduling systems.