AutoOntic:从文本自动构建本体,改进 GraphRAG 知识图谱
AutoOntic: Automatic Ontology Induction from Text for Knowledge Graph Construction
做 GraphRAG 的开发者可以看这篇:很多人用 LLM 抽完实体和关系就得到一张扁平图,没有类型层级,检索质量上不去。AutoOntic 直接从文本自动归纳本体,在 Text2KGBench 上 90% 以上类型语义对齐,MINE 上检索准确率 88%,代码已开源,可以直接接入自己的图谱构建流程试。
论文提出 AutoOntic,一条从文本自动构建本体的流水线,解决 GraphRAG 中无 schema 的 LLM 抽取管道只产出扁平图谱的问题——这类图谱没有类型层级,也没有关系上的 domain/range 约束。其流程分三步:先对开放词表的实体类型和关系做规范化(canonicalization),再在规范化类型上归纳层级,最后挖掘关系级的 domain 和 range 约束。
技术机制上,层级归纳利用语料统计得到的“关系-参数画像”对类型排序,通过局部的、由 LLM 判定的父子关系检查来放置类型,新插入的类型还能触发已有父子链接的重新分配。生成的层级反过来约束实体去重,并支撑关系级约束的归纳。评估方面,论文提出一套参照基准的协议,把归纳出的本体类型分为 exact、broader、narrower、unrelated 四类。
在 Text2KGBench 的 Wikidata-TekGen 和 DBpedia-WebNLG 子集上,使用两个 LLM 后端时,AutoOntic 归纳的端点类型超过 90% 与参照类语义对齐,其中约一半完全匹配,其余差异多在粒度上。在 MINE 基准上,用 AutoOntic 构建的本体组织的知识图谱检索准确率达到 88%。代码已在 GitHub 开源。
AutoOntic: Automatic Ontology Induction from Text for Knowledge Graph Construction
Graph-based retrieval-augmented generation depends on the quality of the knowledge graph (KG) it retrieves over, yet schema-free LLM-based extraction pipelines typically produce flat graphs: deduplicated entities and relations with no type hierarchy and no domain/range constraints. We present AutoOntic, a pipeline that automatically constructs an ontology from text by canonicalizing open-vocabulary entity types and relations, inducing a hierarchy over the canonicalized types, and mining relation-level domain and range constraints. Its hierarchy induction procedure prioritizes types using corpus-derived relation-argument profiles and places them through local, LLM-adjudicated parent/child checks, allowing newly inserted types to trigger reassignment of existing parent-child links. The resulting hierarchy constrains entity deduplication and supports the induction of relation-level domain/range constraints. For intrinsic evaluation, we introduce a reference-based protocol that classifies induced ontology types as exact, broader, narrower, or unrelated to reference domain and range classes. On the Wikidata-TekGen and DBpedia-WebNLG subsets of Text2KGBench, over 90% of AutoOntic's induced endpoint types are semantically aligned with the reference classes across two LLM backbones: about half match exactly, while most remaining differences concern granularity. On MINE, knowledge graphs organized using an ontology built by AutoOntic achieve 88% retrieval accuracy. AutoOntic is available at https://github.com/screemix/AutoOntic.
本条情报归属 AITOP「论文」核心追踪栏目。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:AutoOntic、GraphRAG、知识图谱、本体构建,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.AI」(@Alla Chepurova, Aydar Bulatov, Mikhail Burtsev, Yuri Kuratov),经 AITOP 多维评分管线结构化提炼。