Hallucination, a phenomenon where large language models (LLMs) produce output that is factually incorrect or unrelated to the input, is a major challenge for LLM applications that require accuracy and dependability. In this paper, we introduce a reliable and high-speed production system aimed at detecting and rectifying the hallucination issue within LLMs. Our system encompasses named entity recognition (NER), natural language inference (NLI), span-based detection (SBD), and an intricate decision tree-based process to reliably detect a wide range of hallucinations in LLM responses. Furthermore, our team has crafted a rewriting mechanism that maintains an optimal mix of precision, response time, and cost-effectiveness. We detail the core elements of our framework and underscore the paramount challenges tied to response time, availability, and performance metrics, which are crucial for real-world deployment of these technologies. Our extensive evaluation, utilizing offline data and live production traffic, confirms the efficacy of our proposed framework and service.
View on arXiv@article{wang2025_2407.15441,
title={ Developing a Reliable, Fast, General-Purpose Hallucination Detection and Mitigation Service },
author={ Song Wang and Xun Wang and Jie Mei and Yujia Xie and Sean Muarray and Zhang Li and Lingfeng Wu and Si-Qing Chen and Wayne Xiong },
journal={arXiv preprint arXiv:2407.15441},
year={ 2025 }
}