The World Model Remembers The Actor Forgets Dream

---
name: the-world-model-remembers-the-actor-forgets-dream-
description: 'Skill generated from arXiv paper 2607.19749: The World Model Remembers, the Actor Forgets: Dream Rehearsal for Continual Model-Based RL'
metadata:
  {
    "arxiv": {
      "id": "2607.19749",
      "title": "The World Model Remembers, the Actor Forgets: Dream Rehearsal for Continual Model-Based RL",
      "authors": ['Gurp Nijjer'],
      "published": "2026-07-22",
      "categories": ['cs.LG', 'cs.AI'],
      "url": "https://arxiv.org/abs/2607.19749",
      "utility": 1.0
    }
  }
---
# The World Model Remembers, the Actor Forgets: Dream Rehearsal for Continual Model-Based RL

**arXiv:** 2607.19749  
**Published:** 2026-07-22  
**Authors:** Gurp Nijjer  
**Categories:** cs.LG, cs.AI  
**Utility:** 1.00

## Key Innovation
Model-based reinforcement-learning agents of the DreamerV3 family forget catastrophically when trained on task sequences, even when an unbounded replay buffer preserves every earlier experience. We ask a question the continual-RL literature has assumed an answer to but never measured: which component forgets? Under never-clear replay, pre-registered component-level probes (n=3 seeds throughout) show that the world model retains essentially everything measurable about old tasks -- reward discrimi...

## Potential Application
This paper presents advancements that could be applied to enhance agent capabilities in the areas of cs.LG, cs.AI.

## References
- arXiv: https://arxiv.org/abs/2607.19749