From Symbolic to Neural and Back: Exploring Knowledge Graph-Large Language Model Synergies
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908404312702976 |
|---|---|
| author | Škrlj, Blaž Koloski, Boshko Pollak, Senja Lavrač, Nada |
| author_facet | Škrlj, Blaž Koloski, Boshko Pollak, Senja Lavrač, Nada |
| contents | Integrating structured knowledge from Knowledge Graphs (KGs) into Large Language Models (LLMs) enhances factual grounding and reasoning capabilities. This survey paper systematically examines the synergy between KGs and LLMs, categorizing existing approaches into two main groups: KG-enhanced LLMs, which improve reasoning, reduce hallucinations, and enable complex question answering; and LLM-augmented KGs, which facilitate KG construction, completion, and querying. Through comprehensive analysis, we identify critical gaps and highlight the mutual benefits of structured knowledge integration. Compared to existing surveys, our study uniquely emphasizes scalability, computational efficiency, and data quality. Finally, we propose future research directions, including neuro-symbolic integration, dynamic KG updating, data reliability, and ethical considerations, paving the way for intelligent systems capable of managing more complex real-world knowledge tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_09566 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Symbolic to Neural and Back: Exploring Knowledge Graph-Large Language Model Synergies Škrlj, Blaž Koloski, Boshko Pollak, Senja Lavrač, Nada Computation and Language Artificial Intelligence Machine Learning Integrating structured knowledge from Knowledge Graphs (KGs) into Large Language Models (LLMs) enhances factual grounding and reasoning capabilities. This survey paper systematically examines the synergy between KGs and LLMs, categorizing existing approaches into two main groups: KG-enhanced LLMs, which improve reasoning, reduce hallucinations, and enable complex question answering; and LLM-augmented KGs, which facilitate KG construction, completion, and querying. Through comprehensive analysis, we identify critical gaps and highlight the mutual benefits of structured knowledge integration. Compared to existing surveys, our study uniquely emphasizes scalability, computational efficiency, and data quality. Finally, we propose future research directions, including neuro-symbolic integration, dynamic KG updating, data reliability, and ethical considerations, paving the way for intelligent systems capable of managing more complex real-world knowledge tasks. |
| title | From Symbolic to Neural and Back: Exploring Knowledge Graph-Large Language Model Synergies |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.09566 |