When models disagree on a prediction, it often signals a tricky or risky input...

https://felixssuperperspective.brightsora.com/posts/counterfactual-augmentation-for-disputed-inputs-how-does-it-work

When models disagree on a prediction, it often signals a tricky or risky input worth a closer look. By tracking metrics like ensemble variance or prediction margin, you can spot these uncertain cases