mlsanityData to 14 Oct 2026

Latent Memory Palace: Reasoning for Control as Autoregressive Variational Inference

arXiv 2607.08724 ↗First posted 9 Jul 2026cs.LG cs.ROcited by 0

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

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-texttttok, 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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 22 of its 40 references that are in the map. First posted on arXiv 9 Jul 2026.

Builds on

22 of 40 references.

Closest ancestors · 4

Same direction, newest first

  1. 2026ReGuLaR: Variational Latent Reasoning Guided by Rendered Chain-of-Thought30 Jan
  2. 2025Variational Reasoning for Language Models26 Sep
  3. 2025A Survey on Latent Reasoning8 Jul
  4. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug

Other references · 18

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2013Auto-Encoding Variational Bayescited by 4,053
  3. 2024OpenAI o1 System Cardcited by 1,244
  4. 2024DROID: A Large-Scale In-The-Wild Robot Manipulation Datasetcited by 852
  5. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 797
  6. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 725
  7. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilitiescited by 706
  8. 2026World Action Models are Zero-shot Policiescited by 465
  9. 2024Robotic Control via Embodied Chain-of-Thought Reasoningcited by 346
  10. 2025MolmoAct: Action Reasoning Models that can Reason in Spacecited by 197
  11. 2024Behavior Generation with Latent Actionscited by 160
  12. 2025HAMSTER: Hierarchical Action Models For Open-World Robot Manipulationcited by 130
6 more
  1. 2025A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulationcited by 129
  2. 2025Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokenscited by 124
  3. 2025OneTwoVLA: A Unified Vision-Language-Action Model with Adaptive Reasoningcited by 107
  4. 2025Training Strategies for Efficient Embodied Reasoningcited by 55
  5. 2026Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoningcited by 18
  6. 2025RAVR: Reference-Answer-guided Variational Reasoning for Large Language Modelscited by 8

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Hybrid Latent Reasoning via Reinforcement Learning
  2. 2026LaDi-RL: Latent Diffusion Reasoning Prevents Entropy Collapse in Reinforcement Learning
  3. 2026AutoCRAT: Within-trajectory Joint Control of Stochasticity and Compute for LLM Reasoning
  4. 2025ReLaX: Reasoning with Latent Exploration for Large Reasoning Models
  5. 2026Beyond Test-Time Memory: State-Space Optimal Control for LLM Reasoning
  6. 2026LatentHarness: Learning Latent Actions for Memory and Reasoning via Counterfactual Policy Distillation
  7. 2025The Markovian Thinker: Architecture-Agnostic Linear Scaling of Reasoning
  8. 2026Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents