Tag: distribution-shift

All the articles with the tag "distribution-shift".

• Measuring a Classifier Honestly: Precision, Recall, Thresholds and Calibration

  Updated:19 Oct, 2026

  An 88.5% accurate router that misses 38% of the questions it exists to catch. Why accuracy hides that, how precision, recall and ranking metrics expose it, how to choose a threshold from what each mistake costs without leaking the test set, whether the router's probabilities mean what they say, and what happens when Vietnamese questions arrive.