Continual learning
Ability of a model to acquire new knowledge and skills over time while retaining what it has already learned, rather than losing earlier abilities.
In plain language
The model learns something new without losing what it could already do.
In a sentence
- The continual learning experiment tested both new tasks and earlier ones.
- We evaluated continual learning by checking which older skills the model retained.
Usage note
Acquiring a new skill and retaining earlier skills are both part of the definition. Testing only the new task leaves the retention question unanswered.
See also
- Post-training
Training after pretraining that refines a model's behavior, including following instructions and matching human preferences, and can unlock new skills.
- Knowledge cutoff
Date through which a language model's knowledge base is most extensive and reliable.
Cite this term.
CC BY 4.0, attribution required.
Orbyt Labs, Continual learning (AI Glossary), Nov 14, 2026. https://
Orbyt Labs Glossaries (Nov 14, 2026), https://
Last updated .