Efforts to mitigate hallucinations at the model level include supervised fine-tuning (SFT), reinforcement learning from human feedback (RLHF), contrastive decoding, and grounded pretraining.

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 (3)

• Reinforcement learning from human feedback (RLHF) concept
• supervised fine-tuning concept
• RLHF concept