Large language models may learn erroneous facts with higher confidence than a single-source error would warrant because the internet's tendency to copy and redistribute content creates an amplification dynamic where the model perceives duplicated errors as consensus.

Authors

• Person: M. Brenndoerfer Organization: mbrenndoerfer.com
  Hallucination Causes: Why Language Models Fabricate Facts

Sources

• Hallucination Causes: Why Language Models Fabricate Facts mbrenndoerfer.com M. Brenndoerfer · mbrenndoerfer.com via serper

Referenced by nodes (1)

• Large Language Models concept