Helen Ngo

Mitigating harm in language models with conditional-likelihood filtration

2021 · arXiv (Cornell University)

Language models trained on large-scale unfiltered datasets curated from the open web acquire systemic biases, prejudices, and harmful views from their training data. We present a methodology for programmatically identifying and removing harmful text from …

Performance of a Screening Mammography AI Algorithm Repurposed for Symptomatic Mammography in a Tertiary Outpatient Clinic

2026 · Diagnostics

Background/Objectives: The aim of the study was to evaluate the diagnostic accuracy of a commercial artificial intelligence (AI) algorithm originally developed for screening mammography when applied to symptomatic women presenting to a tertiary outpatient clinic. …