Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2407.15441 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909556521566208 |
|---|---|
| author | Wang, Song Wang, Xun Mei, Jie Xie, Yujia Muarray, Sean Li, Zhang Wu, Lingfeng Chen, Si-Qing Xiong, Wayne |
| author_facet | Wang, Song Wang, Xun Mei, Jie Xie, Yujia Muarray, Sean Li, Zhang Wu, Lingfeng Chen, Si-Qing Xiong, Wayne |
| contents | 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, we have 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_15441 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Developing a Reliable, Fast, General-Purpose Hallucination Detection and Mitigation Service Wang, Song Wang, Xun Mei, Jie Xie, Yujia Muarray, Sean Li, Zhang Wu, Lingfeng Chen, Si-Qing Xiong, Wayne Computation and Language 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, we have 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. |
| title | Developing a Reliable, Fast, General-Purpose Hallucination Detection and Mitigation Service |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2407.15441 |