Intrinsic factors within model architecture, training data quality, and sampling algorithms significantly contribute to hallucination problems in large language models.

Authors

• Person: Not available Organization: Frontiers
  Survey and analysis of hallucinations in large language models

Sources

• Survey and analysis of hallucinations in large language models www.frontiersin.org Frontiers via serper

Referenced by nodes (2)

• Large Language Models concept
• hallucination concept