procedure
Self-refinement mitigates LLM hallucinations by having the model review and adjust its own output before presenting the final response.

Authors

Sources

Referenced by nodes (1)