Improving Reliability of Large Language Models for Nuclear Power Plant Diagnostics [Poster]
- Idaho National Laboratory (INL), Idaho Falls, ID (United States)
Large Language Models (LLMs) struggle out of the box when answering factually about detailed questions, especially in domains that are sparsely represented in their training data. This causes hallucinations and reduces reliability making it difficult for them to be used in practice. This work shows that using RAG techniques can improve factual accuracy and reliability, allowing for the application of LLMs in specialized areas, even when those areas that aren’t extensively covered in their initial training.
- Research Organization:
- Idaho National Laboratory (INL), Idaho Falls, ID (United States)
- Sponsoring Organization:
- USDOE Office of Nuclear Energy (NE)
- DOE Contract Number:
- AC07-05ID14517
- OSTI ID:
- 2440146
- Report Number(s):
- INL/EXP--24-79591-Rev000
- Country of Publication:
- United States
- Language:
- English
Similar Records
Improving Reliability of Large Language Models for Nuclear Power Plant Diagnostics Technical Presentation
Reducing AI RAG Hallucination by Optimizing Routing Techniques
Automating and Evaluating Large Language Models for Accurate Text Summarization Under Zero-Shot Conditions
Conference
·
Wed Aug 07 00:00:00 EDT 2024
·
OSTI ID:2440149
Reducing AI RAG Hallucination by Optimizing Routing Techniques
Conference
·
Fri Aug 16 00:00:00 EDT 2024
·
OSTI ID:2474834
Automating and Evaluating Large Language Models for Accurate Text Summarization Under Zero-Shot Conditions
Conference
·
Fri Feb 28 23:00:00 EST 2025
·
OSTI ID:2538344