keyword

human uncertainty

Human uncertainty is the inherent variability, ambiguity, and lack of consensus among human judgments when interpreting, evaluating, or categorizing information. In computational fields such as machine learning and data science, it represents the natural divergence in how different individuals perceive subjective, complex, or under-specified inputs rather than simple annotator error. Because multiple valid perspectives can exist simultaneously, human uncertainty reflects situations where a single deterministic ground truth is insufficient, often leading practitioners to represent human responses as statistical distributions, class frequencies, or entropy measures across a population of annotators.

1 item

Stop Measuring Calibration When Humans Disagree

Joris Baan, Wilker Aziz, Barbara Plank, Raquel Fernández

OrganizationsIT University of CopenhagenLudwig Maximilian University of MunichMunich Center for Machine LearningUniversity of Amsterdam

Why you should read this

Demonstrates the theoretical flaws of standard calibration metrics when evaluated against majority labels on ambiguous tasks and proposes instance-level measures that directly align model confidence with full distributions of human judgment.

Calibration is a popular framework to evaluate whether a classifier knows when it does not know—i.e., its predictive probabilities are a good indication of how likely a prediction is to be correct. Correctness is commonly estimated against the human majority class. Recently, calibration to human majority has been measured on tasks where humans inherently disagree about which class applies. We show that measuring calibration to human majority given inherent disagreements is theoretically problematic, demonstrate this empirically on the ChaosNLI dataset, and derive several instance-level measures of calibration that capture key statistical properties of human judgements—class frequency, ranking and entropy.1

Source

https://aclanthology.org/2022.emnlp-main.124.pdf

Added

2026-10-02