Agents manage when to check facts to save resources
When Should Agents Check External State? Budgeting Observations for Stored Intentions
Artificial Intelligence
Summary
Some agents remember tasks they need to do when certain things happen later, but checking whether those things have happened can be costly. The authors introduce BudgetPM, a way for agents to decide when and how often to check for these future triggers while staying within a limited budget of observations. Their system learns policies that smartly choose when to spend or save checking capacity, improving performance with fewer checks. This helps agents balance current and future needs to use their resources better.
What this means in practice
- •For virtual assistant developers: Optimize task checking schedules in virtual assistants to reduce costly data retrieval while maintaining decision quality.
- •For automated monitoring system engineers: Improve resource use in monitoring systems by allocating observation budgets across multiple conditions effectively.
Authors
Zhengkun Di, Bin Shi, Kai Sun, Yiming Xu, Bo Dong
Abstract
Prospective memory allows an agent to retain an intention tied to a future condition, but the stored intention does not reveal whether that condition currently holds. Checking it may require web access, multi-step tool use, and paid calls. Existing systems decide when intentions require attention, but do not allocate the resulting observations under a shared budget. We introduce the first resource-allocation formulation for the external observations required by stored intentions under a shared episode budget. BudgetPM offers two policy variants that share a hard-budget executor. BudgetPM-Static uses a lightweight Logistic scorer to learn whether a check improves the current decision. BudgetPM-Sequential distills full-episode hindsight schedules into a lightweight policy that decides when to spend or reserve capacity using only pre-query information at deployment. We evaluate BudgetPM against two public memory-agent systems, five matched controls, and four hand-designed monitoring or budget-adaptation rules. Across two benchmarks and three backbones, BudgetPM-Static outperforms adapted Mem0 and PMA workflows. On PM-Bench, its Logistic scorer reaches competitive quality--cost operating points alongside higher-capacity scorers and retains 99.9--100\% of unconstrained quality with 42--54\% fewer observations. Under severe scarcity and the same hard caps, BudgetPM-Sequential exceeds the strongest tested natural monitoring schedule by 1.92--2.58 Set F1 points. It reaches the same Set F1 and on-time recall with 16--33\% fewer observations. Matched attribution, exact-cost analysis, and a fixed-budget load intervention link this gain to competition between present and future opportunities. These results yield a demand--capacity design rule: local gating works when capacity covers demand, while future-aware supervision adds value when observations compete across time.