research
A World Model Should Forget What Stopped Being True
Differential retention separates permanent invariants from instance facts that should be revised quickly.
Summary
Differential retention separates permanent invariants from instance facts that should be revised quickly.
A position paper argues that ordinary continual-learning scores misclassify useful revision as catastrophic forgetting when the environment itself changes. It proposes testing invariant knowledge, such as physics and object permanence, separately from revision latency for mutable facts. The metric is a framework proposal rather than evidence from a new trained model.
Why it matters
Differential retention separates permanent invariants from instance facts that should be revised quickly.
Limits and context
No additional limitation was separately recorded.
Key claims
Differential retention separates permanent invariants from instance facts that should be revised quickly.
Evidence: source-2026-10-05-004
Sources
- arXiv preprint 2610.03713arXiv · primary research
Corrections
No corrections have been recorded for this story.