AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore60

What Should World Models Forget? Stratified Retention for Continual Adaptation

Summary

This study argues that continual learning wrongly treats any performance drop as forgetting, and that world models should retain knowledge in layers, forgetting selectively as they adapt.

Why this matters

Covers model · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items