CADOC：面向长时程智能体的缓存感知动态对象上下文

中文摘要 AI 辅助

对于长时程智能体而言，上下文是瓶颈：每次请求都会重新发送历史记录，窗口限制了任务长度，并且随着历史增长，推理能力会退化。用紧凑的检索卡片替换结构化对象可以缩短提示词，并保留可检索的精确原始内容，但编辑历史可能会破坏前缀缓存复用，而先前的可恢复方法通过预测未来复用或预设间隔来安排编辑时机。我们提出CADOC（缓存感知动态对象上下文），这是一种在线算法，用紧凑卡片替换结构化对象，同时保留其原始内容的精确按需检索。CADOC通过权衡累积等待成本与共享缓存重建成本，批量调度替换操作。其调度规则遵循经济订货量权衡，在平稳假设下恢复最优整数批次。在评估中，CADOC始终在比较配置中实现最低的总输入成本，平均降低约40%的输入成本，同时保持接近完整上下文的任务性能。因此，CADOC提供了一种基于成本的可压缩上下文管理方法，表明高效压缩不仅取决于缩短提示词，还取决于调度编辑以保留缓存复用。

英文摘要

For a long-horizon agent, context is the bottleneck: the history is resent with every request, the window caps task length, and reasoning degrades as the history grows. Replacing structured objects with compact retrieval Cards shortens the prompt and keeps the exact originals retrievable, but editing the history can break prefix-cache reuse, and prior recoverable methods time their edits by forecasts of future reuse or by preset intervals. We propose CADOC (Cache-Aware Dynamic Object Context), an online algorithm that replaces structured objects with compact Cards while preserving exact, on-demand retrieval of their original contents. CADOC schedules replacements in batches by balancing accumulated waiting cost against shared cache-reconstruction cost. Its scheduling rule follows from an economic order quantity trade-off, recovers the optimal integer batch under stationary assumptions. Across evaluation, CADOC consistently achieves the lowest aggregate input cost among the compared configurations, which reduces input cost by approximately 40\% on average while maintaining task performance close to full context. CADOC thus provides a cost-derived approach to compressible context management, demonstrating that efficient compression depends not only on shortening prompts but also on scheduling edits to preserve cache reuse.

发表机构

• Shanghai Jiao Tong University（上海交通大学）
• Institute of Natural Sciences, Shanghai Jiao Tong University（上海交通大学自然科学研究院）

机构由 AI 辅助整理，请以论文原文为准。