Benchmarking LLMs via Uncertainty Quantification

arXiv:2401.12794v2 Announce Type: replace
Abstract: The proliferation of open-source Large Language Models (LLMs) from various institutions has highlighted the urgent need for comprehensive evaluation methods. However, current evaluation platforms, such as the widely recognized HuggingFace open LLM leaderboard, neglect a crucial aspect -- uncertainty, which is vital for thoroughly assessing LLMs. To bridge this gap, we introduce a new benchmarking approach for LLMs that integrates uncertainty quantification. Our examination involves eight LLMs (LLM series) spanning five representative natural language processing …

abstract arxiv benchmarking bridge cs.cl current evaluation gap however huggingface language language models large language large language models leaderboard llm llm leaderboard llms open llm leaderboard platforms quantification type uncertainty via vital