EvoOntology：面向数据智能体的自演化本体层

AI 摘要

该论文提出 EvoOntology，一个面向数据智能体的自演化本体层，旨在弥合智能体与异构数据（表格、文件、数据库）之间的鸿沟。EvoOntology 将本体封装为包含模式层、内容层和工具层的 MCP 服务器，使智能体可在运行时主动查询本体，并通过构建智能体与自演化循环持续优化本体。在三个数据智能体基准和四个 LLM 基座上的实验表明，其性能持续优于强基线及现有语义层方法。

正文节选

EvoOntology: A Self-Evolving Ontology Layer for Data Agents Abstract Data agents aim to fulfill natural-language instructions over heterogeneous data, including tables, files, and databases. However, data agents face a challenging agent-data gap: heterogeneous data resides outside the agent, while the agent can access it (e.g., column names and file paths) only through generic tools. Existing approaches either let agents directly explore raw data sources or inject manually constructed semantic l

发布时间：—

抓取时间：2026-09-21 12:43

来源机构：Hugging Face