AgentLoop：面向工具增强型LLM智能体的槽闭合执行循环的运行时控制
AgentLoop: Runtime Control of Slot-closed Execution Loops for Tool-augmented LLM Agents
浏览论文内容
中文总结 AI 辅助
针对工具增强型LLM智能体缺乏任务完成显式信号的问题，提出AgentLoop，通过槽闭合与状态驱动控制实现运行时循环管理，显著降低令牌成本与服务调用次数。
中文摘要 AI 辅助
工具增强型大语言模型（LLM）智能体正成为服务计算中的重要执行单元，但现有智能体循环仍缺乏用于评估任务完成情况的显式运行时信号。挑战在于，即使运行时上下文已停止变化，智能体仍可能继续推理或调用服务，而已收集的证据仍未综合成完整答案，这导致资源使用效率低下。为应对这些挑战，本文提出AgentLoop，为工具增强型智能体提供槽闭合执行循环的运行时控制。槽闭合意味着请求所需的信息槽已被足够的运行时证据覆盖，且在循环停止前明确识别未解决的槽。AgentLoop将开放式智能体迭代转换为状态驱动的执行控制：它维护紧凑的运行时状态，使用模型辅助的结构化验证来检查答案完整性和缺失证据，并在相邻的LLM/工具轮次上应用有界稳定性和低增益信号，然后选择三种动作之一：继续调用、答案综合或终止迭代。实验表明，与基线相比，AgentLoop减少了冗余执行和上下文增长，总令牌成本最多降低88.44%，平均服务调用次数最多减少76.85%。消融研究进一步表明，以槽为中心的控��路径发挥核心作用，因为禁用它会增加执行深度并大幅降低准确性。总体而言，结果表明高效的工具增强型智能体可受益于显式运行时信号，以决定何时进一步的LLM/工具迭代不再增加有用上下文或支持证据。
英文摘要
Tool-augmented large language model (LLM) agents are becoming an important execution unit in service computing, but existing agent loops still lack explicit runtime signals for assessing task completion. The challenge lies in the fact that an agent may continue reasoning or invoking services even after the runtime context has stopped changing, while evidence already collected remains unsynthesized into a complete answer, which leads to inefficiency in resource usage. To address these challenges, this paper presents AgentLoop, which provides runtime control of slot-closed execution loops for tool-augmented agents. Slot closure means that the information slots required by a request have been covered by sufficient runtime evidence, and that unresolved slots are explicitly identified before the loop stops. AgentLoop converts open-ended agent iteration into state-driven execution control: it maintains a compact runtime state, uses model-assisted structured verification to check answer completeness and missing evidence, and applies bounded stability and low-gain signals over neighboring LLM/tool rounds before selecting one of three actions: Continue Invocation, Answer Synthesis, or Terminate Iteration. Experiments show that AgentLoop reduces redundant execution and context growth, with total token cost reduced by up to 88.44% and average service invocations reduced by up to 76.85% against baselines. The ablation study further shows that the slot-centered control path plays a central role, since disabling it increases execution depth and substantially reduces accuracy. Overall, the results suggest that efficient tool-augmented agents can benefit from explicit runtime signals for deciding when further LLM/tool iterations no longer add useful context or supported evidence.
发表机构
- Southern University of Science and Technology（南方科技大学）
- Shenzhen University of Advanced Technology（深圳先进技术大学）
- Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）
- Institute of AI and Brain Sciences, Department of Computer Science, University of Macau（澳门大学计算机科学学院人工智能与脑科学研究所）
机构由 AI 辅助整理，请以论文原文为准。
