Knowledge Tree
entity alignment
Facts (17)
Sources
LLM-empowered knowledge graph construction: A survey - arXiv arxiv.org Oct 23, 2025 7 facts
claimGraph-level models, such as those proposed by Liu et al. (2022), integrate semantic cues to enhance the robustness of entity alignment in Knowledge Fusion.
referenceClassical approaches to entity alignment in Knowledge Fusion relied on lexical and structural similarity measures, as reviewed in Zeng et al. (2021).
claimMulti-feature fusion strategies for Knowledge Fusion combine structural, attribute, and relational similarities to improve entity alignment, as demonstrated by Yang et al. (2022a).
referenceZeng et al. (2021) authored 'A comprehensive survey of entity alignment for knowledge graphs', published in the journal AI Open, volume 2, pages 1–13.
claimEmbedding-based techniques for entity alignment, which align entities within shared vector spaces, have improved scalability and automation in Knowledge Fusion, as surveyed by Zhu et al. (2024a).
referenceLLM-Align (Chen et al., 2024) treats entity alignment as a constrained multiple-choice problem using multi-step prompting to enhance semantic discrimination.
claimInstance-level fusion in knowledge graphs aims to reconcile heterogeneous or redundant entities through entity alignment, disambiguation, deduplication, and conflict resolution to maintain a coherent and semantically precise graph.
Practices, opportunities and challenges in the fusion of knowledge ... frontiersin.org 6 facts
referenceChen et al. (2024) proposed a method for entity alignment using noisy annotations generated by large language models, as described in arXiv preprint arXiv:2405.16806.
claimConstructing and maintaining high-quality knowledge graphs typically involves significant human effort, including data cleaning, entity alignment, relation labeling, and expert validation, which is particularly labor-intensive in domains requiring expert knowledge.
referenceAutoAlign (Zhang R. et al., 2023) performs entity alignment by constructing a predicate proximity graph to capture predicate similarity between Knowledge Graphs and uses the TransE model (Bordes et al., 2013) to compute entity embeddings, aligning entities into a shared vector space.
referenceLLM-Align (Chen X. et al., 2024) performs entity alignment by selecting important entity attributes and relations via heuristic methods, inputting entity triples into a Large Language Model (LLM) to infer alignment results, and using a multi-round voting mechanism to mitigate hallucinations and positional bias.
referenceThe paper 'Two heads are better than one: Integrating knowledge from knowledge graphs and large language models for entity alignment' was published as an arXiv preprint (arXiv:2401.16960) in 2024.
referenceThe paper 'Llm-align: utilizing large language models for entity alignment in knowledge graphs' (arXiv:2412.04690) investigates the use of large language models for entity alignment tasks within knowledge graphs.
Knowledge Graphs: Opportunities and Challenges - Springer Nature link.springer.com Apr 3, 2023 4 facts
claimEntity alignment or ontology alignment is the primary method of knowledge fusion, aiming to match the same entity across multiple knowledge graphs.
claimEntity alignment methods that only consider single-modality knowledge graphs perform poorly because they fail to fully reflect the relationships of entities as they exist in the real world.
referenceMao X, Wang W, Xu H et al. published 'Mraea: an efficient and robust entity alignment approach for cross-lingual knowledge graph' in the proceedings of the 13th International Conference on Web Search and Data Mining in 2020.
claimEntity alignment is currently the primary method used for implementing knowledge fusion tasks.