微软提出 FOCUS 方法在测试时压缩智能体上下文
微软一篇新论文提出 FOCUS 方法，在测试时压缩智能体上下文：它判断智能体的下一次决策实际依赖哪些历史交互，保留这些单元并丢弃其余部分。该方法无需训练数据或微调，可作为独立层置于闭源 API 模型之前；在工具调用、QA、网页和多轮对话基准上，相比使用完整历史，峰值上下文最多减少 48%，任务成功率最多提升 8.9 分。论文地址：academy.dair.ai/papers/focus…
New paper from Microsoft on compressing agent context at test time.
If your agent gets worse as its interaction history grows, this method is worth a look.
FOCUS asks which past interactions the agent's next decisions actually depend on. It keeps those units of the history and drops the others. It needs no training data or fine-tuning, so it works as a separate layer in front of closed-API models.
On tool-calling, QA, web and multi-turn dialogue benchmarks, it cuts peak context by up to 48% and raises task success by up to 8.9 points compared with running on the full history.
Paper: academy.dair.ai/papers/focus…
来源：dair_ai · x.com