§ AI Wiki / Glossary
One-line definitions, the AI dictionary.
§ Search this category
Search the Wiki →Empirical relationships describing how model performance changes with parameters, data, and compute.
Systematic skews in a model's outputs that favor certain groups or viewpoints, usually inherited from training data or design choices.
A machine learning paradigm in which the model generates its own training signal from unlabeled data.
The initial training phase where a model learns general language ability from trillions of tokens of generic data.
The general term for how a model picks the next token from its probability distribution.
An agent's ability to detect errors or failed steps and automatically recover from them.
A technique in which a model evaluates its own output against explicit criteria to surface errors and weaknesses.
A practical chunking strategy that splits documents on coarse separators first, then progressively finer ones.
A model type that generates the next token step-by-step, conditioned on previous tokens.
A mechanism where each element in a sequence attends to every other element in the same sequence.