亚马逊微软论文：长程智能体无需每步调用大模型
New Amazon Microsoft paper shows long-horizon agents should not need an LLM deci…
Agent 架构的重要优化思路，通过去中心化决策显著降低 LLM 调用成本并提升长程任务成功率，值得 Agent 开发者关注其技术细节。
New Amazon Microsoft paper shows long-horizon agents should not need an LLM decision after every tiny action; the hard part is knowing which actions can safely run together.
最新一篇来自亚马逊和微软的论文表明，长视界智能体不应在每次微小动作后都需要大语言模型进行决策；难点在于判断哪些动作可以安全地并行执行。
SPACE learns those boundaries from successful trajectories.
SPACE 从成功轨迹中学习这些边界。
It converts trajectories into programmatic skills, treats subskill boundaries as labels for meaningful chunks, then distills them into a policy that emits variable-length primitive actions with no skill library at test time.
它将轨迹转化为程序化技能，将子技能边界视为有意义片段的标签，然后将其蒸馏为一个策略，该策略在测试时不依赖技能库即可输出变长的原始动作。
On ScienceWorld, success rose from 35.9% to 67.2%, while average LLM rounds fell from 10.2 to 5.2.
在 ScienceWorld 上，成功率从 35.9% 提升至 67.2%，而平均大语言模型轮次从 10.2 降至 5.2。
do not make agents reconsider every tiny step, and do not blindly batch actions either. Train them to learn when to keep acting and when to look again.
不要让智能体对每个微小步骤都重新考虑，也不要盲目地批量执行动作。训练它们学会何时继续行动，何时再次审视。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力