The max_new_tokens parameter controls sequence length in large language models, and longer generations face higher cumulative exposure bias divergence, which increases hallucination risk as the sequence grows.

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 (3)

• Large Language Models concept
• hallucination concept
• exposure bias concept