AgentLoop: Runtime Control of Slot-closed Execution Loops for Tool-augmented LLM Agents

2026-09-27

AI Summary

This study addresses the problem of ineffective iterations and resource waste in tool-augmented large language model (LLM) agents caused by the absence of runtime signals. To this end, it proposes AgentLoop, a framework that innovatively introduces a slot-closure mechanism to transform open-ended iteration into state-driven execution. By leveraging bounded stability signals, AgentLoop dynamically determines whether to continue, synthesize, or terminate actions. It further achieves efficient runtime control through model-assisted structured verification, compact state maintenance, and low-gain signal detection. Experimental results demonstrate that AgentLoop reduces total token costs by 88.44% and average service invocations by 76.85%, substantially improving agent execution efficiency.

Abstract

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.

Problem

• Tool-augmented LLM agents
• Task completion assessment
• Runtime control
• Resource inefficiency
• Agent execution loops

Innovation

• Slot-closed Execution
• Runtime Control
• Tool-augmented LLM Agents
• State-driven Execution
• Structured Verification