Hallucination Rate from Dataset Contamination Calculator

Estimate AI model hallucination rates based on training data contamination analysis. Model the relationship between unseen fact density, retrieval quality, and factual accuracy across different knowledge domains

Each component has a specific meaning:

• Training data contamination analysis — The training data contamination analysis recorded for the patient or scenario being assessed.

Note: Interpret the hallucination rate result against the clinical thresholds and context described above.

Enter the training data contamination analysis for the patient or scenario you are assessing. Estimate AI model hallucination rates based on training data contamination analysis. Model the relationship between unseen fact density, retrieval quality, and factual accuracy across different knowledge domains. Use the hallucination rate result to inform your clinical assessment.