# ARC-Bench: Closed-Loop Replanning Masks Broken Action Ranking in Frozen JEPA World Models

**Published:** September 9, 2026
**Author:** Zhengshu Zhang, Zhiyuan Li
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/arc-bench-closed-loop-replanning-masks-broken-action-ranking-in-frozen-j

arXiv:2609.05461v1 Announce Type: new Abstract: Reward-free latent world models plan by scoring candidate actions with distances in a frozen latent space: an action is preferred if its predicted future embedding lands closer to the goal embedding. This silently assumes that…

> From the original source: **arXiv — cs.AI**
> [Read full article at arXiv — cs.AI](https://arxiv.org/abs/2609.05461)

**Tags:** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*