这篇论文用实验告诉你，深的世界模型在滚动预测时不一定管用——在DeepMind Control的9个任务里，有2个反而是早退浅层更好，而且原因竟然出在训练方式上。搞模型规划的值得看看。
这篇论文在9个DeepMind Control任务上引入shallow penalty ρ来测量深度在自回归展开中的效果，发现6/9的任务中深度帮助（ρ最高达4.7倍），2/9的任务中浅层胜出（ρ低至0.85倍）。通过消融实验，反转现象（cheetah任务）被证实由训练目标造成：仅监督早期退出第一步时ρ从0.87升至1.18，而内在折中任务不受影响。ρ的部分可预测性通过观察/动作维度和单步模型误差与ρ的Spearman相关系数约0.75（n=9）展示。在CEM规划器中，ρ的符号可预测规划是否受益于深度，反转任务上浅层规划优于深层。
When Does Depth Survive Composition? Compute--Quality Regimes in Latent World Models
Adaptive-compute world models -- early-exit or mixture-of-depths predictors that spend variable depth per step -- assume depth buys better predictions and can be routed adaptively. In autoregressive rollouts, the first assumption requires depth's per-step precision to survive composition. We test this with a pre-registered instrument, the shallow penalty $ρ=\mathrm{err}(\text{shallowest-exit rollout})/\mathrm{err}(\text{full-depth rollout})$, across nine DeepMind Control tasks under matched single-step ($K=1$) and multi-step ($K=4$) training, three seeds each. We find three regimes: on 6/9 tasks depth helps rollouts (intrinsic, $ρ$ up to $4.7\times$), on 2/9 the shallow exits beat the full stack (inversion, $ρ$ down to $0.85\times$), and one is flat. The robust inversion (cheetah) is not a property of the dynamics but is created by training: an ablation supervising early exits only at the first rollout step erases it ($ρ: 0.87\to1.18$, $n=8$, $Δ=+0.31$), while an intrinsic-tradeoff task is unaffected -- a double dissociation we call the routability catch-22, since the supervision that makes exits routable is what trains them to out-roll the full stack. The regime is partly predictable a priori: observation/action dimensionality and one-step model error correlate with $ρ$ at $|\text{Spearman}|\approx0.75$ ($n=9$). Inside a CEM planner, $ρ$'s sign predicts whether planning benefits from depth, most sharply on the inversion task, where shallow planning beats deep. Finally, three cautions: a task's regime depends on the metric space, the rollout horizon, and the encoder. All thresholds and gates were fixed before the compute campaign, including a pre-registered negative for the hypothesis that motivated the study.