Year: 2024
Type: article
Abstract: Large Language Models (LLMs) like GPT have demonstrated impressive capabilities in generating human-like text.However, their tendency to produce hallucinations-plausible-sounding but factually incorrect responses-poses significant challenges for their deployment in critical applications.This review paper synthesizes various strategies to mitigate hallucinations in GPT models, including prompt engineering, finetuning, retrieval augmentation, and truth triangulation.By combining these methods, we aim to enhance the reliability and accuracy of LLM outputs.This paper provides a comprehensive overview of current approaches, their effectiveness, and potential future directions in reducing AI hallucinations.
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4401624777