Back to researchers
Sang Michael Xie
Holistic evaluation of language models (HELM)
Co-authored HELM: a framework for evaluating language models across many axes beyond raw accuracy.
Highlights
EvaluationBenchmarksResponsible AI
Focus: Holistic evaluation of language models (HELM)
Why it matters: Co-authored HELM: a framework for evaluating language models across many axes beyond raw accuracy.
Research Areas
EvaluationBenchmarksResponsible AI