Skip to content

Holistic Evaluation of Language Models

Authors: Liang et al. Year: 2023 ArXiv/Link: https://arxiv.org/abs/2211.09110

Summary

Comprehensive evaluation framework (HELM) covering diverse dimensions of language model behavior and capabilities.

Key Concepts

  • Evaluation framework
  • Comprehensive assessment
  • Multiple dimensions
  • Benchmark diversity
  • Model reliability

Impact

Established comprehensive evaluation methodology for LLMs

Category

Safety & Evaluation