Emergent Ability
A complex capability that spontaneously appears in a machine learning model only after it reaches a critical threshold of scale in parameters or training data.
Think of It Like This
Like water turning into ice: the temperature drops gradually with no visible change until a specific point is reached and a new state suddenly forms.
Researchers have observed that large language models sometimes develop unexpected skills—such as advanced arithmetic, reasoning, or translation—without explicit training. These abilities are virtually nonexistent in smaller models and unpredictably manifest as compute scales up. Understanding whether these are true emergent phenomena or measurement artifacts remains heavily debated.