Learning to Wait: Synchronizing Agents with the Physical World

TL;DR

This paper introduces an agent-side method enabling large language models to synchronize with real-world timing by predicting wait durations, improving efficiency and temporal coordination in asynchronous environments.

Contribution

It extends the Code-as-Action paradigm to the temporal domain, allowing LLMs to predict precise waiting times without environment-side modifications.

Findings

Agents accurately predict wait durations to synchronize with environment

Method reduces query overhead and execution latency

Temporal awareness is shown to be a learnable skill for LLMs

Abstract

Real-world agentic tasks, unlike synchronous Markov Decision Processes (MDPs), often involve non-blocking actions with variable latencies, creating a fundamental \textit{Temporal Gap} between action initiation and completion. Existing environment-side solutions, such as blocking wrappers or frequent polling, either limit scalability or dilute the agent's context window with redundant observations. In this work, we propose an \textbf{Agent-side Approach} that empowers Large Language Models (LLMs) to actively align their \textit{Cognitive Timeline} with the physical world. By extending the Code-as-Action paradigm to the temporal domain, agents utilize semantic priors and In-Context Learning (ICL) to predict precise waiting durations (\texttt{time.sleep(t)}), effectively synchronizing with asynchronous environment without exhaustive checking. Experiments in a simulated Kubernetes cluster demonstrate that agents can precisely calibrate their internal clocks to minimize both query overhead and execution latency, validating that temporal awareness is a learnable capability essential for autonomous evolution in open-ended environments.