<p class="jove_content">We propose a reinforcement-learning-based fine-tuning approach for large language models that uses the Enti&#8239;Hallucination&#8239;Index (EHI) as a reward signal to reduce entity-level hallucination in text summarization. Experiments on meeting transcripts show that this method improves entity faithfulness and accuracy.</p>

fine-tuning-large-language-models-using-entity-hallucination-index

Fine-Tuning Large Language Models Using Entity Hallucination Index for Text Summarization

Recent advancements in large language models (LLMs) have led to notable improvements in abstractive summarization quality. However, hallucination - ...

Institution - Informatica Business Solutions

Articles Published in JoVE