Anticipate and Learn: Unleashing Idle-Time Compute in Proactive Agents
2026-05-25 • Computation and Language
Computation and LanguageInformation RetrievalMultiagent Systems
AI summaryⓘ
The authors explain that current AI agents only respond after being asked, missing chances to prepare for what users might need next. They developed ProAct, an AI that uses the waiting time between user interactions to predict and gather information for future questions. Their tests show ProAct helps users finish tasks faster, with less effort, and makes fewer mistakes compared to regular AI agents. They also created ProActEval, a big set of tests to check how well proactive AI works across many situations.
AI agentsproactive computingdialogue historypersistent memoryknowledge gapstask completionuser efforthallucination ratebenchmarkreflective accuracy
Authors
Haoyi Hu, Qirong Lyu, Xianghan Kong, Weiwen Liu, Jianghao Lin, Zixuan Guo, Yan Xu, Yasheng Wang, Weinan Zhang, Yong Yu
Abstract
While AI agents demonstrate remarkable capabilities in reasoning and tool use, they remain fundamentally reactive: they compute responses only after explicit user prompts. This paradigm ignores a critical opportunity: the idle time between interactions is largely wasted, leaving agents unable to prepare for future user needs. To bridge this gap, we introduce ProAct, a proactive agent architecture that leverages idle-time compute to anticipate and fulfill likely upcoming user needs. By analyzing evolving dialogue history together with persistent memory, ProAct predicts upcoming needs and iteratively acquires information, allowing the agent to resolve knowledge gaps and prepare evidence before the user initiates a query.To rigorously evaluate proactive capabilities, we also introduce ProActEval, a comprehensive benchmark comprising 200 scenarios across 40 domains, featuring predictable need chains and diverse user cognitive profiles. Empirical results demonstrate significant advantages over reactive baselines. ProAct accelerates task completion by reducing required turns by 14.8%, decreases user effort by 11.7%, and cuts hallucination rates by 28.1% on ProActEval. Furthermore, MemBench evaluations confirm that ProAct achieves state-of-the-art reflective accuracy, underscoring its sustained and robust performance.