做长任务 Agent 开发的团队终于有了一个不碰模型权重就能提升性能的方案——AdaCoM 用一个小模型当上下文管家，实测搜索任务提升 39%，值得在项目里试试。
一篇新论文提出 AdaCoM，通过一个独立的小模型来清理和组织 Agent 的上下文，从而提升其在长任务中的表现，无需重新训练 Agent 本身。AdaCoM 在 Agent 每一步行动前，对任务历史进行重写、合并、剪枝或保留，然后让原始 Agent 基于清理后的上下文行动。与简单摘要不同，AdaCoM 能学习不同 Agent 需要何种上下文——强 Agent 可保留更多原始历史，弱 Agent 则需要更简洁的笔记。在网页搜索和深度研究任务上，AdaCoM 将平均搜索性能提升了 39%。
AI agent can get better at long tasks without retr…
AI agent can get better at long tasks without retraining the agent itself, by using a separate small model to clean and organize its context.
Moves context management outside the agent, so a separate helper can clean up the task history while the main agent stays unchanged.
The paper proposes AdaCoM, which is a separate LLM that edits the agent’s working context before the agent takes its next step.
AdaCoM places a separate, trained manager between the task history and the frozen agent, so the agent does not need to learn a new memory habit or expose its weights.
Before each step, this manager can rewrite, merge, prune, or preserve parts of the running context, then the original agent acts on the cleaned version.
That sounds like summarization, but the distinction matters.
A summary assumes the right answer is compression, while AdaCoM learns that different agents need different kinds of context to stay competent, because stronger agents can use more raw history while weaker agents need shorter and cleaner notes.
They tested AdaCoM on web search and deep research tasks across several agents, and it improved average web search performance by 39%.
----
Link – arxiv. org/abs/2605.30785
Title: "Learning Agent-Compatible Context Management for Long-Horizon Tasks"