Librarian Agents 论文：让智能体自行整理长上下文以提升检索效果
How do we scale up the context of LLMs?
Chelsea Finn 转发 Anikait Singh 等人的新论文 Librarian Agents，研究智能体能否通过自行整理长上下文工作区、预判查询来适应专业领域的大量非结构化上下文。她指出智能体搜索可扩展到任意规模的信息但容易漏掉相关内容，而针对下游用途优化信息组织在 OOD 领域也能带来 30+% 的提升。推文附有论文和代码链接。
查看原帖部署在专业领域（从企业代码库到律师事务所）的大语言模型，要在堆积如山的非结构化上下文中进行挖掘！ 我们的新论文《Librarian Agents》提出：智能体能否通过自行整理其长上下文工作区、在查询到来之前预判查询，从而实现自适应？🧵⬇️
查看原文
LLMs deployed in specialized domains, from enterprise codebases to law firms, dig through a mountain of unstructured context! Our new paper, Librarian Agents, asks: can an agent adapt by curating its own long-context workspace, anticipating queries before they arrive? 🧵⬇️
正文
当前显示原文，中文全文译文尚未就绪。
How do we scale up the context of LLMs?
Agentic search scales to arbitrarily large amounts of info, but can easily miss relevant info.
Idea💡: Optimize organization of info for downstream utility. Leads to 30+% gains even on OOD domains