Calibration as a First-Class Criterion in LLM Evaluation

arXiv:2609.26489 · cs.CL · Submitted 2026-09-22 · Read on arXiv

cs.CL

Submitted: 2026-09-22

Updated: 2026-09-22

Comments: Accepted to the 3rd Workshop on Uncertainty-Aware NLP (UncertaiNLP) at EMNLP 2026

License: http://creativecommons.org/licenses/by/4.0/

The gist: Calibration of language models -- the alignment between expressed or implicit confidence and empirical correctness -- is a well-studied subfield within NLP.

Terminology

Abstract

Calibration of language models -- the alignment between expressed or implicit confidence and empirical correctness -- is a well-studied subfield within NLP. Methods to measure it already exist. The problem is adoption: outside this subfield, NLP research regularly introduces new models, datasets, and benchmarks without checking whether the model's confidence scores are meaningful. We argue that this adoption gap is a major obstacle to trustworthy LLM evaluation. Miscalibration causes problems in two distinct areas: at deployment, where overconfident mistakes cause real harm, and inside the research pipeline, where methods like LLM-as-a-judge, synthetic data generation, and active learning rely on calibrated confidence without verifying it. Standard calibration metrics only require two inputs per example: a confidence score and a correctness judgment. Most benchmarks in use today already provide both, meaning calibration can be reported immediately. For open-ended generation, however, defining these two inputs is still an open challenge. We argue that each NLP subfield should pair its main performance metric with a calibration score and call for treating calibration as an essential property of every model rather than a niche topic.

Sources

Related papers