liveBook · Manning

Robustness

As explained in Evaluating AI Systems

The ability of a system to maintain suitable behavior under varied, unclear, or challenging inputs and conditions.

A robust evaluation metric gives consistent, meaningful results when minor, irrelevant factors change. A fragile metric can swing sharply because of noise or superficial differences, making results less trustworthy and progress harder to track. section 3.3.5

When small changes distort scores

BLEU may drop when a synonym or punctuation mark changes despite equivalent meaning. Perplexity can shift with tokenization or preprocessing, while human ratings can vary with wording or presentation order. section 3.3.5 section 3.3.5

Check whether a metric remains stable by:

  • Testing controlled variations that should not affect measurement, such as paraphrased patient questions or changed response formatting. section 3.3.5
  • Repeating evaluation across datasets, domains, or environments to see whether results hold. section 3.3.5
  • Resampling evaluation data to estimate how much the metric varies across samples. section 3.3.5

How Evaluating AI Systems covers it

How deep it goes
Defines metric robustness, contrasts fragile metrics, and outlines perturbation, cross-environment, and resampling checks.
Examples
Uses BLEU, perplexity, human ratings, and medical-chatbot input variations to illustrate metric sensitivity.
What you can do after
Evaluate metric stability by perturbing inputs, comparing environments, and resampling evaluation data.
How it connects
Connects metric robustness to AI evaluation, evaluation metrics, and quality dimensions such as fairness and explainability.

Also explained in

Compare how 12 books explain robustness →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Evaluating AI Systems. The section links point to the parts of the book it draws on.