CORVUS: Context Optimization and Reduction Via Underlying Synchronization for LLM Coding Agents

AI 中文解读

AI编程代理终于摆脱了“记性差”的毛病。传统方式下，AI在处理代码时会不断把文件快照一股脑塞进历史记录，结果文件被改动后这些旧快照就成了累赘，导致AI犯迷糊、反复读文件，白白浪费算力。CORVUS提出的新架构就像给AI装了个“实时文件同步器”，它不再死记硬背每个读取瞬间的内容，而是维护一份文件最新状态的注册表，每次思考时自动拉取当前实际内容。这样一来，AI不再做重复劳动，推理负担骤降。实验显示，输入token平均减少9%到50%，最终提示文本缩短15%到32%，推理步骤最多减少37%，而解题成功率基本不变。对普通用户来说，这意味着用GitHub Copilot这类AI编程工具时，等待时间会明显缩短，每月算力消耗也相应减少——未来AI编程服务更可能降价或免费。企业开发者则能实时、准确地处理多人协作的代码库，避免AI因读到过期文件而给出错误建议。从更广的视角看，这种“轻装上阵”的架构设计思路，可能被推广到其他需要处理动态信息的AI场景，比如智能文档生成或实时数据分析，让AI不再被自己的“回忆”拖后腿。

arXiv:2607.22711v1 Announce Type: new Abstract: LLM coding agents operate by constructing trajectories that accumulate reasoning, tool calls, and results to enable multi-step decision-making. However, the conventional append-only trajectory architecture found in practice tightly couples file-read actions with their observations, capturing snapshots that become permanently fixed in the chronological history. As files change through agent edits or concurrent human modifications, these snapshots become stale, causing reasoning errors and causing agents to redundantly re-read files, with each re-read appending yet another copy to the trajectory. To mitigate this, we propose CORVUS, a novel trajectory architecture that decouples file-read actions from their observations by maintaining a synchronized registry of relevant files and injecting only their current contents at each reasoning cycle. This structural change produces significantly lighter-weight trajectories that remain synchronized with the actual codebase state by construction, eliminating redundant file copies and stale snapshots that bloat conventional trajectories. We evaluated CORVUS on SWE- POLYBENCH_VERIFIED and SWE-BENCH PRO across four LLMs, achieving 9-50% reduction in average input tokens per task, 15-32% shorter final prompts, and up to 37% fewer reasoning cycles while maintaining comparable pass rates.