HuDEx: Integrating Hallucination Detection and Explainability for Enhancing the Reliability of LLM responses
Sujeong Lee, Hayoung Lee, Seongsoo Heo, Wonik Choi
TL;DR
HuDEx addresses the critical challenge of hallucinations in LLMs by jointly detecting hallucinations and generating explanations to improve reliability. It leverages a compact, PEFT-tuned Llama-based backbone (LoRA) trained on diverse detection datasets and augmented with explanation data, achieving superior detection accuracy compared to larger models across multiple benchmarks. The framework also evaluates explanations with a dedicated LLM judge, showing competitive factuality and clarity, and demonstrates zero-shot robustness. While effective, HuDEx relies on internal knowledge when sources are unavailable, suggesting future enhancements via external retrieval and automated feedback to further boost reliability and explainability in real-world deployments.
Abstract
Recent advances in large language models (LLMs) have shown promising improvements, often surpassing existing methods across a wide range of downstream tasks in natural language processing. However, these models still face challenges, which may hinder their practical applicability. For example, the phenomenon of hallucination is known to compromise the reliability of LLMs, especially in fields that demand high factual precision. Current benchmarks primarily focus on hallucination detection and factuality evaluation but do not extend beyond identification. This paper proposes an explanation enhanced hallucination-detection model, coined as HuDEx, aimed at enhancing the reliability of LLM-generated responses by both detecting hallucinations and providing detailed explanations. The proposed model provides a novel approach to integrate detection with explanations, and enable both users and the LLM itself to understand and reduce errors. Our measurement results demonstrate that the proposed model surpasses larger LLMs, such as Llama3 70B and GPT-4, in hallucination detection accuracy, while maintaining reliable explanations. Furthermore, the proposed model performs well in both zero-shot and other test environments, showcasing its adaptability across diverse benchmark datasets. The proposed approach further enhances the hallucination detection research by introducing a novel approach to integrating interpretability with hallucination detection, which further enhances the performance and reliability of evaluating hallucinations in language models.