Competency Questions as Executable Plans: a Controlled RAG Architecture for Cultural Heritage Storytelling

arXiv cs.AI / 4/6/2026

💬 OpinionIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper proposes a neuro-symbolic, knowledge-graph-based controlled RAG architecture for cultural heritage storytelling to reduce LLM hallucinations and improve factual veracity.
It repurposes competency questions (CQs) as run-time executable narrative plans in a transparent plan–retrieve–generate workflow that is evidence-closed and auditable.
The approach uses a new resource, the Live Aid KG, which multimodally aligns 1985 concert data with an ontology and links external multimedia assets to support richer, verifiable narratives.
The authors compare three RAG strategies over the graph—symbolic KG-RAG, text-enriched Hybrid-RAG, and structure-aware Graph-RAG—and report a measurable trade-off among factual precision, contextual richness, and narrative coherence.
The results provide design guidance for building personalised and controllable storytelling systems for domains where correctness and traceability are critical.

Abstract

The preservation of intangible cultural heritage is a critical challenge as collective memory fades over time. While Large Language Models (LLMs) offer a promising avenue for generating engaging narratives, their propensity for factual inaccuracies or "hallucinations" makes them unreliable for heritage applications where veracity is a central requirement. To address this, we propose a novel neuro-symbolic architecture grounded in Knowledge Graphs (KGs) that establishes a transparent "plan-retrieve-generate" workflow for story generation. A key novelty of our approach is the repurposing of competency questions (CQs) - traditionally design-time validation artifacts - into run-time executable narrative plans. This approach bridges the gap between high-level user personas and atomic knowledge retrieval, ensuring that generation is evidence-closed and fully auditable. We validate this architecture using a new resource: the Live Aid KG, a multimodal dataset aligning 1985 concert data with the Music Meta Ontology and linking to external multimedia assets. We present a systematic comparative evaluation of three distinct Retrieval-Augmented Generation (RAG) strategies over this graph: a purely symbolic KG-RAG, a text-enriched Hybrid-RAG, and a structure-aware Graph-RAG. Our experiments reveal a quantifiable trade-off between the factual precision of symbolic retrieval, the contextual richness of hybrid methods, and the narrative coherence of graph-based traversal. Our findings offer actionable insights for designing personalised and controllable storytelling systems.