Are Large Language Models Effective Knowledge Graph Constructors?

Chen, R., Jiang, W., Qin, C., & Xiong, B. (2025). Are Large Language Models Effective Knowledge Graph Constructors?.

Type

Preprint

Intellectual base

Grey literature

Year

2025

Link

https://arxiv.org/abs/2510.11297

Cited from

mathematics

Cited by

1 citation across 1 artifact.

Each citation links to the sentence it supports in the citing article.

Primes

• Connectedness
  • Linguistics and knowledge graphs. Terminologies, ontologies, and citation networks are connected when no concept is an orphan unreachable from the rest, and orphan-detection is a connectedness audit.

    This sourceEvaluates LLM-constructed knowledge graphs from structural and semantic perspectives; structural connectivity (giant-component coverage, isolated/orphan nodes) is treated as a quality metric, since graph-based retrieval and multi-hop reasoning rely on reachability and degrade with disconnected components — orphan-detection is a connectedness audit over the concept graph.

Verification

This reference passed the adversarial substantiation pipeline: it was checked to exist and to support the claim it is attached to. See how references were verified.

Registry ID ref:b9642b4040aa · see in the full table