Enhancing Metacognitive AI: Knowledge-Graph Population with Graph-Theoretic LLM Enrichment
Quick Take
MetaKGEnrich enhances LLMs with automated knowledge repair through knowledge graph population and self-diagnosis.
Key Points
- Automated pipeline for self-directed knowledge repair.
- Improved answer quality in 80-87% of tested queries.
- Utilizes graph metrics and targeted retrieval methods.
📖 Reader Mode
~2 min readAbstract:Metacognition-the ability to monitor one's own knowledge state, spot gaps, and autonomously fill them--remains largely absent from modern AI. Here, we present MetaKGEnrich, a fully automated pipeline that endows large language model (LLM) applications with self-directed knowledge repair. The system (i) builds knowledge graphs from a seed query, (ii) detects sparse regions via seven graph metrics, (iii) has GPT-4o generate targeted questions, (iv) retrieves web evidence with Tavily and ingests it into Neo4j, and (v) re-answers the query with GraphRAG for GPT-4 to evaluate improvement. Tested on 30 queries from each of three widely-used datasets: Google Research Natural Questions, MS MARCO, and Hot-potQA. MetaKGEnrich improved answer quality in 80% of HotpotQA questions, 87% of Google Research Natural Questions and 83% of MS MARCO questions, while preserving well-supported regions. This proof of concept demonstrates how topological self-diagnosis plus targeted retrieval can advance AI toward humanlike metacognitive learning.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.16676 [cs.AI] |
| (or arXiv:2605.16676v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2605.16676 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Brendan Conway-Smith [view email]
[v1]
Fri, 15 May 2026 22:32:07 UTC (762 KB)
— Originally published at arxiv.org
More from arXiv cs.AI
See more →From Prompts to Protocols: An AI Agent for Laboratory Automation
An AI agent integrates large language models for automating laboratory protocols, enhancing efficiency and accuracy.