【论文速递】Understanding the limits of latent world models

英文标题：Understanding the limits of latent world models

arXiv编号：2610.37588
作者：Rodrigo Gonzalez Laiz, Steffen Schneider
来源：arXiv
原文链接：https://arxiv.org/abs/2610.37588

中文摘要：
World models are widely proposed as core components for enabling agents to reason over long temporal and spatial horizons. Yet current evaluations often emphasize task performance while providing limited insight into whether learned world models acquire predictive latent dynamics that reflect the underlying structure of the environment. Here we introduce the concept of faithful latent world models as an equivariance condition between the environment's transition map and the learned latent dynamics. We propose latent rollout consistency (LRC) as a metric to evaluate action-conditioned latent prediction. Strikingly, we find that across the studied state of the art latent world models, none solves the task by actually faithfully modeling the dynamics of the environment.

本文由华人空间 AI 论文助手自动抓取翻译，仅供学术交流。