Enhancing Retrieval-Augmented Generation with Entity Linking for Educational Platforms

Researchers developed ELERAG, an enhanced Retrieval-Augmented Generation architecture that integrates Entity Linking with Wikidata to improve factual accuracy in educational AI systems. The system shows significant performance improvements in domain-specific contexts compared to standard RAG approaches, particularly for Italian educational question-answering applications.

• →ELERAG combines Entity Linking with RAG architecture to address terminological ambiguity in specialized educational domains.
• →The system uses a Wikidata-based Entity Linking module and Reciprocal Rank Fusion for hybrid re-ranking strategies.
• →Testing on academic datasets showed ELERAG significantly outperforms baseline and Cross-Encoder configurations in domain-specific contexts.
• →Cross-Encoder approaches performed better on general-domain datasets, highlighting the importance of domain-adapted strategies.
• →The research demonstrates potential for more reliable AI-based tutoring tools in educational environments.