Addressing Hallucinations with RAG and NMISS in Italian Healthcare LLM Chatbots

TL;DR

This paper presents a combined detection and mitigation approach using RAG and NMISS to reduce hallucinations in Italian healthcare LLM chatbots, improving answer accuracy and contextual relevance.

Contribution

It introduces NMISS for better hallucination detection and demonstrates how RAG and NMISS together enhance LLM reliability in healthcare applications.

Findings

GPT-4 outperforms other models in accuracy

NMISS benefits mid-tier models significantly

Combined approach reduces hallucination occurrences

Abstract

I combine detection and mitigation techniques to addresses hallucinations in Large Language Models (LLMs). Mitigation is achieved in a question-answering Retrieval-Augmented Generation (RAG) framework while detection is obtained by introducing the Negative Missing Information Scoring System (NMISS), which accounts for contextual relevance in responses. While RAG mitigates hallucinations by grounding answers in external data, NMISS refines the evaluation by identifying cases where traditional metrics incorrectly flag contextually accurate responses as hallucinations. I use Italian health news articles as context to evaluate LLM performance. Results show that Gemma2 and GPT-4 outperform the other models, with GPT-4 producing answers closely aligned with reference responses. Mid-tier models, such as Llama2, Llama3, and Mistral benefit significantly from NMISS, highlighting their ability to provide richer contextual information. This combined approach offers new insights into the reduction and more accurate assessment of hallucinations in LLMs, with applications in real-world healthcare tasks and other domains.