general

Definition

A claim about breadth: the method or model is meant to apply across tasks and domains rather than being tuned for one. It signals a value judgement — coverage without retraining is treated as more important than topping a single leaderboard. Distinguish it from unified, which describes one shared mechanism handling several things; a system can be unified in architecture yet still narrow in what it is good at.

Explainers using this term

• Self-Improving AI — Self-Play, Co-Evolution, and Generated CurriculaMastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm