Jia-Nian Lin

Mitigating Hallucinations in Natural Language Generation through Prompt Engineering: A Mechanism- Oriented Narrative Review

Large language models (LLMs) can generate fluent and confident responses that are factually incorrect, unsupported by evidence, or inconsistent with the source material. These hallucinations reduce the reliability of LLM-based question answering, summariza tion, dialogue, and information retrieval systems, especially w...