Competency Questions as Executable Plans: a Controlled RAG Architecture for Cultural Heritage Storytelling
📰 ArXiv cs.AI
A controlled RAG architecture uses competency questions as executable plans to generate accurate cultural heritage stories with Large Language Models
Action Steps
- Define competency questions to guide the narrative generation process
- Integrate Knowledge Graphs with Large Language Models to ensure factual accuracy
- Implement a controlled RAG architecture to execute plans and generate stories
- Evaluate the generated stories for accuracy and coherence
Who Needs to Know This
AI engineers and researchers on a team can benefit from this approach to improve the accuracy of LLM-generated narratives, while cultural heritage preservationists can leverage this technology to create engaging and reliable stories
Key Insight
💡 Using competency questions as executable plans can improve the accuracy of LLM-generated narratives in cultural heritage applications
Share This
📚💡 Controlled RAG architecture for accurate cultural heritage storytelling with LLMs
Key Takeaways
A controlled RAG architecture uses competency questions as executable plans to generate accurate cultural heritage stories with Large Language Models
Full Article
Title: Competency Questions as Executable Plans: a Controlled RAG Architecture for Cultural Heritage Storytelling
Abstract:
arXiv:2604.02545v1 Announce Type: new 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 (
Abstract:
arXiv:2604.02545v1 Announce Type: new 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 (
DeepCamp AI