RoleRAG: Enhancing LLM Role-Playing via Graph Guided Retrieval
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_ | 1866915303262257152 |
|---|---|
| author | Wang, Yongjie Leung, Jonathan Shen, Zhiqi |
| author_facet | Wang, Yongjie Leung, Jonathan Shen, Zhiqi |
| contents | Large Language Models (LLMs) have shown promise in character imitation, enabling immersive and engaging conversations. However, they often generate content that is irrelevant or inconsistent with a character's background. We attribute these failures to: (1) the inability to accurately recall character-specific knowledge due to entity ambiguity, and (2) a lack of awareness of the character's cognitive boundaries. To address these issues, we propose RoleRAG, a retrieval-based framework that integrates efficient entity disambiguation for knowledge indexing with a boundary-aware retriever for extracting contextually appropriate information from a structured knowledge graph. Experiments on role-playing benchmarks show that RoleRAG's calibrated retrieval helps both general-purpose and role-specific LLMs better align with character knowledge and reduce hallucinated responses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_18541 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RoleRAG: Enhancing LLM Role-Playing via Graph Guided Retrieval Wang, Yongjie Leung, Jonathan Shen, Zhiqi Artificial Intelligence 68T50 I.2.7 Large Language Models (LLMs) have shown promise in character imitation, enabling immersive and engaging conversations. However, they often generate content that is irrelevant or inconsistent with a character's background. We attribute these failures to: (1) the inability to accurately recall character-specific knowledge due to entity ambiguity, and (2) a lack of awareness of the character's cognitive boundaries. To address these issues, we propose RoleRAG, a retrieval-based framework that integrates efficient entity disambiguation for knowledge indexing with a boundary-aware retriever for extracting contextually appropriate information from a structured knowledge graph. Experiments on role-playing benchmarks show that RoleRAG's calibrated retrieval helps both general-purpose and role-specific LLMs better align with character knowledge and reduce hallucinated responses. |
| title | RoleRAG: Enhancing LLM Role-Playing via Graph Guided Retrieval |
| topic | Artificial Intelligence 68T50 I.2.7 |
| url | https://arxiv.org/abs/2505.18541 |