arXiv cs.MA· Nick Hagar·· 9 小时前AI 评分34
FieldGuide:面向 token 高效智能体文档探索的语料地图
FieldGuide: A Corpus Map for Token-Efficient Agentic Document Exploration
AI 导读
FieldGuide 是一个面向 AI 智能体的语料地图,将文档集合组织为主题层级,并提供语料定位、范围搜索和受限阅读工具。在 38 份芝加哥警察局公开记录文档的 5 项新闻文档探索任务中,它比原始文件访问少用 27% token、41% 时间,比 RAG 少用 38% token、21% 时间。准确性审计仅发现 4 处轻微问题,但盲评生成的报道线索评分低于原始文件访问。
正文 · 原文
Abstract:Generative AI agents can help journalists explore large document collections. But exploratory workflows can require substantial token use and time, especially when the task is broad and the relevant subquestions are not yet known. This paper introduces FieldGuide, an agent-facing corpus map that organizes a document collection into a topic hierarchy and exposes tools for corpus orientation, scoped search, and bounded reading. I evaluate FieldGuide on five journalistic document-exploration tasks using 38 Chicago Police Department public-records documents, comparing it with raw-file access and retrieval-augmented generation. Across the five tasks, FieldGuide used 27% fewer tokens and 41% less elapsed time than raw-file access; relative to RAG, it used 38% fewer tokens and 21% less elapsed time. An accuracy audit identified four minor issues in FieldGuide's responses, comparable to the other conditions. In blind ratings of generated reporting leads, FieldGuide's mean ratings were higher than RAG's and lower than the raw-file condition's. These descriptive results suggest that corpus orientation may reduce runtime token use and elapsed time without severe accuracy failures. They also motivate further study of whether selective access limits broad, cross-document discovery.
| Comments: | Presented at Computation + Journalism Symposium 2026 |
| Subjects: | Computation and Language (cs.CL); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2611.21647 [cs.CL] |
| (or arXiv:2611.21647v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.21647 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Nick Hagar [view email]
[v1]
Wed, 14 Oct 2026 17:34:19 UTC (75 KB)
来源:arXiv cs.MA · arxiv.org