Enhancing Retrieval-Augmented Generation With Entity Linking For Educational Platforms

Francesco Granata, University of Catania

Francesco Poggi, National Research Council

Misael Mongiovì, University of Catania

Date

12/2025

AI Overview

Why it matters

Large language models can hallucinate incorrect information in critical educational domains. Standard retrieval-augmented generation relies on semantic similarity, which often fails with specialized terminology, polysemous terms, and instructional vocabulary.

Study at a glance

Question: Can combining entity linking with semantic search improve the factual accuracy of educational question-answering systems?
Setting and data: Italian telematic university course lectures (Applied Economics, Language and Communication; 50 classes, 32 hours), transcribed into 676 chunks.
Method: Model and benchmark evaluation comparing different re-ranking and retrieval pipelines against baseline architectures.
AI tools: ELERAG, GPT-4o, Whisper Turbo, multilingual-e5-large, mmarco-mMiniLMv2-L12-H384-v1

Key findings

On the educational benchmark, the proposed entity-linking system achieved the highest Exact Match (0.565) and Precision@1 (0.696), outperforming both the standard baseline and cross-encoders.
On the general-domain Wikipedia benchmark, the cross-encoder performed best with an Exact Match of about 0.777, demonstrating a domain mismatch effect between generic and specialized settings.
In full pipeline generation, the proposed entity-linking architecture yielded the highest Exact Match (0.551) and precision (0.458), showing that improved retrieval ranking translates to better model outputs.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

In the era of Large Language Models (LLMs), Retrieval-Augmented Generation (RAG) architectures are gaining significant attention for their ability to ground language generation in reliable knowledge sources. Despite their impressive effectiveness in many areas, RAG systems based solely on semantic similarity often fail to ensure factual accuracy in specialized domains, where terminological ambiguity can affect retrieval relevance. This study proposes an enhanced RAG architecture that integrates a factual signal derived from Entity Linking to improve the accuracy of educational question-answering systems in Italian. The system includes a Wikidata-based Entity Linking module and implements three re-ranking strategies to combine semantic and entity-based information: a hybrid score weighting model, reciprocal rank fusion, and a cross-encoder re-ranker. Experiments were conducted on two benchmarks: a custom academic dataset and the standard SQuAD-it dataset. Results show that, in domain-specific contexts, the hybrid schema based on reciprocal rank fusion significantly outperforms both the baseline and the cross-encoder approach, while the cross-encoder achieves the best results on the general-domain dataset. These findings confirm the presence of an effect of domain mismatch and highlight the importance of domain adaptation and hybrid ranking strategies to enhance factual precision and reliability in retrieval-augmented generation. They also demonstrate the potential of entity-aware RAG systems in educational environments, fostering adaptive and reliable AI-based tutoring tools.