将网页操作任务编译为高效执行代码，显著降低延迟并提升准确性。
Agent JIT Compilation for Latency-Optimizing Web Agent Planning and Scheduling
- 把自然语言任务直接编译成带并行优化的可执行代码。
- 相比基线模型提速10.4倍，准确率提升28%。
- 适合追求低延迟与高可靠性的自动化网页操作场景。
计算机使用代理（CUAs）通过生成浏览器点击、输入、滚动等工具调用序列，自动执行自然语言描述的任务，如“订购塔可贝尔最便宜的商品”。现有实现采用逐轮获取屏幕截图-执行-调用LLM的循环，导致高延迟和因工具误用引发的频繁错误。本文提出代理即时编译（JIT-Compilation），将任务描述直接编译为包含LLM调用、工具调用和并行化策略的可执行代码。系统包含三部分：(1) JIT-Planner，生成多个代码方案，基于工具规范验证并选择成本最低者；(2) JIT-Scheduler，通过蒙特卡洛成本估计探索并行策略，利用学习到的延迟分布；(3) 不变性保障工具协议，规定前置与后置条件，减少工具误用。在五个应用中，JIT-Planner相较Browser-Use实现10.4倍加速与28%准确率提升，JIT-Scheduler相较OpenAI CUA实现2.4倍加速与9%准确率提升。
原文摘要 · Abstract (English)
Computer-use agents (CUAs) automate tasks specified with natural language such as "order the cheapest item from Taco Bell" by generating sequences of calls to tools such as click, type, and scroll on a browser. Current implementations follow a sequential fetch-screenshot-execute loop where each iteration requires an LLM call, resulting in high latency and frequent errors from incorrect tool use. We present agent just-in-time (JIT) compilation, a system that compiles task descriptions directly into executable code that may include LLM calls, tool calls, and parallelization. Our approach comprises three components: (1) JIT-Planner, which generates multiple code plans, validates each against tool specifications, and selects the minimum-cost candidate; (2) JIT-Scheduler, which explores parallelization strategies via Monte Carlo cost estimation from learned latency distributions; and (3) an invariant-enforcing tool protocol specifying precondition and postcondition requirements to reduce the rate of incorrect tool use. Across five applications, JIT-Planner achieves $10.4\times$ speedup and 28$\%$ higher accuracy over Browser-Use, while JIT-Scheduler achieves $2.4\times$ speedup and 9\% higher accuracy over OpenAI CUA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。