English

A Framework for Evaluating LLMs Under Task Indeterminacy

Machine Learning 2024-11-22 v1 Computation and Language Human-Computer Interaction

Abstract

Large language model (LLM) evaluations often assume there is a single correct response -- a gold label -- for each item in the evaluation corpus. However, some tasks can be ambiguous -- i.e., they provide insufficient information to identify a unique interpretation -- or vague -- i.e., they do not clearly indicate where to draw the line when making a determination. Both ambiguity and vagueness can cause task indeterminacy -- the condition where some items in the evaluation corpus have more than one correct response. In this paper, we develop a framework for evaluating LLMs under task indeterminacy. Our framework disentangles the relationships between task specification, human ratings, and LLM responses in the LLM evaluation pipeline. Using our framework, we conduct a synthetic experiment showing that evaluations that use the "gold label" assumption underestimate the true performance. We also provide a method for estimating an error-adjusted performance interval given partial knowledge about indeterminate items in the evaluation corpus. We conclude by outlining implications of our work for the research community.

Keywords

Cite

@article{arxiv.2411.13760,
  title  = {A Framework for Evaluating LLMs Under Task Indeterminacy},
  author = {Luke Guerdan and Hanna Wallach and Solon Barocas and Alexandra Chouldechova},
  journal= {arXiv preprint arXiv:2411.13760},
  year   = {2024}
}

Comments

To Appear in NeurIPS 2024 Workshops on Evaluating Evaluations (EvalEval) and Statistical Foundations of LLMs and Foundation Models (SFLLM)

R2 v1 2026-06-28T20:07:13.757Z