Rigorous Interpretation Is a Form of Evaluation
Abstract
Current machine learning models are evaluated through behavioral snapshots, with benchmark accuracies, win rates and outcome-based metrics. Model explanations and evaluations, however, are fundamentally intertwined: understanding why a model produces a behavior can be as important as measuring what it produces. If we trusted interpretability, we argue that it can serve not merely as diagnostics but as a richer and more principled form of model evaluation beyond surface-level performance metrics. We explore three ways interpretability can function evaluatively: (1) fixing problems by identifying the root causes of unwanted behavior, (2) detecting subtly faulty mechanisms that invalidate model outputs, and (3) predicting potential issues before they arise by fully understanding the model's weaknesses. To fulfill its evaluative potential, we argue that interpretability methods must generate claims that are falsifiable, reproducible, and predictive -- that is, interpretability must meet scientific standards.
Cite
@article{arxiv.2605.05508,
title = {Rigorous Interpretation Is a Form of Evaluation},
author = {Isabelle Lee and Emmy Liu and Cathy Jiao and Brihi Joshi and Dani Yogatama and Fazl Barez and Michael Saxon},
journal= {arXiv preprint arXiv:2605.05508},
year = {2026}
}