Hendrycks, 2021 — Aligning {AI} With Shared Human Values

DOI: <a href="https://doi.org/10.48550/arXiv.2008.07172" target="_blank" rel="noopener noreferrer">https://doi.org/10.48550/arXiv.2008.07172
URL: <a href="https://arxiv.org/abs/2008.07172" target="_blank" rel="noopener noreferrer">https://arxiv.org/abs/2008.07172

ETHICS benchmark and dataset for aligning models with shared human moral judgments.

Publication links

ETHICS benchmark and dataset for aligning models with shared human moral judgments.