Granularity Is a Decision: Joint Horizon–Description Selection in Latent World Models

Embodied agents increasingly plan with world models, rolling candidate actions forward and executing the one whose predicted outcome best reaches the goal. Such models are typically trained as Markov decision process (MDP) transitions $s_{t+1}\sim P(\cdot \mid s_t,a_t)$, binding every prediction to one environment step and one state representation. However, many physical interactions are action-sparse with persistent effects: a single throw sends an object through flight, collision, and settling, a cascade that unfolds without further action yet decides success. Step-wise prediction of such cascades compounds error, coarse horizons skip decisive contacts, and full detail wastes capacity. All three costs stem from one fixed choice of granularity, made once at design time. Thus, we propose making granularity a variable that the world model chooses per prediction. We formalize the idea as an MDP while every model query carries a prediction request $r=(\Delta ,\alpha )$ that advances the imagined rollout $\Delta$ steps at description level $\alpha$. Across three physics benchmarks, granularity proves consequential: predicting in long jumps cuts compounding rollout error by orders of magnitude, yet no fixed request suits every test set: the one that best separates successful from failed launches on two sets ranks worst on a third. That variance is exploitable: choosing the request per prediction beats a fixed request tuned on other test sets, and choosing both axes matches a horizon-only adaptive baseline at a third of its compute. These results establish granularity as a decision variable rather than a design constant: consequential, unsolvable by any fixed setting, and cheap to adapt.