In attention mechanisms, the ability to focus on relevant parts of the input sequence while ignoring others is often referred to as _____.

In attention mechanisms the ability to highlight important parts of an input sequence while down‑weighting irrelevant parts is called attention. This works by giving each input element a score that reflects how useful it is for the current prediction. The scores are turned into weights that emphasize the useful elements and diminish the rest, so the model “looks” where it needs to. For example, when translating a sentence, the model can give a high weight to the word “apple” when predicting the French word “pomme” and a low weight to the filler word “the. ” This selective weighting is what lets attention models perform well on many language tasks.