cvs-health/uqlm: UQLM: Uncertainty Quantification for Language Models, is a Python package for UQ-based LLM hallucination detection

UQLM is a Python package designed for detecting hallucinations in Large Language Models using uncertainty quantification techniques. It features four types of scorers: black-box, white-box, LLM-as-a-judge, and ensemble. Each scorer provides a confidence score indicating the likelihood of errors in LLM outputs. Users can install the package via PyPI and access various example notebooks for implementation guidance.