智能体路由器：一种基于执行的带记忆的持续学习方法

中文摘要 AI 辅助

大型语言模型（LLM）智能体为基于命令行的网络操作提供了颇具前景的接口，但合理的命令在执行后仍可能失败或引入操作风险。现有方法主要关注命令生成或最终配置的正确性，未利用基于执行的经验来共同提升候选覆盖度与动作选择。我们提出一种基于执行的双路径后果感知智能体，用于基于CLI的SONiC操作，该智能体生成多个完整动作，预测其执行后果，并通过效用与风险感知的重排序选择最终动作。提案侧路径将可复用的操作经验抽象为可检索的指导，以提升可行动作覆盖度且无需修改提案LLM；选择侧路径通过会话级LoRA更新，利用真实SSH反馈调整后果预测器，以提升条件选择质量。在使用不同Qwen3提案模型的多轮SONiC操作会话实验中，该框架提升了可行动作覆盖度与Top-1执行成功率，且两条自适应路径在交互中提供互补增益。

英文摘要

Large language model (LLM) agents provide a promising interface for command-line-based network operations, but a plausible command may still fail or introduce operational risk after execution. Existing approaches mainly focus on command generation or final configuration correctness, and do not use execution-grounded experience to jointly improve candidate coverage and action selection. We propose an execution-grounded dual-path consequence-aware agent for CLI-based SONiC operations, which generates multiple complete actions, predicts their execution consequences, and selects the final action through utility- and risk-aware reranking. The proposal-side path abstracts reusable operational lessons into retrievable guidance to improve feasible-action coverage without modifying the proposal LLM, while the selection-side path adapts the consequence predictor through session-level LoRA updates using real SSH feedback to improve conditional selection quality. Experiments over multi-turn SONiC operation sessions with different Qwen3 proposal models show that the framework improves feasible-action coverage and top-1 execution success, and that the two adaptation paths provide complementary gains over interaction.

发表机构

• Zhejiang University（浙江大学）
• Zhejiang Lab（之江实验室）
• State Grid Hebei Electric Power Co., Ltd.（国网河北省电力有限公司）
• Macau University of Science and Technology（澳门科技大学）

机构由 AI 辅助整理，请以论文原文为准。