这篇论文提出了Latent Memory Palace方法，把推理当成自回归变分推断来控制机器人，在模拟和现实任务都很强，还带可变长动作分词器，做控制方向的朋友可以看看。
Latent Memory Palace (LMP) 方法提出将控制策略的推理形式化为自回归变分推断，通过组织信息于潜在记忆宫殿实现自适应迭代检索。该方法推导出潜在空间强化学习技术以优化变分下界，得到策略 LMP-π 在仿真和真实场景中表现优异，并展示可解释的测试时计算分配。LMP-tok 作为变长动作分词器，显著提升了自回归策略的性能。论文提供了通过变分推断视角进行潜在推理控制的新观点。
Latent Memory Palace: Reasoning for Control as Autoregressive Variational Inference
Human decision-making is highly flexible -- some actions are taken immediately; others require longer deliberation. Language models have exhibited a similar capacity for adaptive "reasoning." However, transferring this capability to continuous control policies has been challenging, as directly reasoning in language space may lack the granularity for spatial understanding and precise motions. In this work, we show that reasoning for control policies can emerge by organizing information in an autoregressive latent space reminiscent of a memory palace, where retrieval is iterative and adaptive. Our method, Latent Memory Palace (LMP), formulates reasoning as variational inference with an autoregressive latent distribution. We derive a latent-space reinforcement learning technique to tractably optimize its variational lower bound. The resulting policy, LMP-$π$, achieves strong empirical performance in simulation and real-world domains while exhibiting interpretable, adaptive allocation of test-time compute. We further show that the same framework yields a variable-length action tokenizer, LMP-$\texttt{tok}$, which significantly improves the performance of downstream autoregressive policies. Together, these results present a new perspective on latent reasoning for control through the lens of variational inference.