2025
Beyond the Safety Bundle: Auditing the Helpful and Harmless Dataset
NAACL 2025long
In an effort to mitigate the harms of large language models (LLMs), learning from human feedback (LHF) has been used to steer LLMs towards outputs that are intended to be both less harmful and more helpful. Despite the widespread adoption of LHF in practice, the quality of this feedback and its effe…