Reinforcement Learning from Human Feedback (RLHF) aligns large language model behavior with factual correctness, but has low feasibility due to complex setup requirements.

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 (1)

• Reinforcement learning from human feedback (RLHF) concept