mlsanityData to 14 Oct 2026

Think Fast, Plan Selectively: Adaptive Deliberation for Efficient Data-Driven MPC

arXiv 2609.32591 ↗First posted 26 Sep 2026cs.RO cs.LGcited by 0

On the map: World models → Latent dynamics models and model-based RL

Abstract (arXiv)

Data-driven model predictive control (MPC) combines learned world models with online trajectory optimization, achieving strong performance in continuous control. However, the per-step cost of sampling and evaluating hundreds of candidate trajectories restricts deployment to control frequencies well below what real-time robotics demands. Motivated by the dual-process theory of human cognition, which distinguishes between fast, intuitive processing (System 1) and slower, deliberative reasoning (System 2), we ask whether every decision requires the same degree of computational deliberation. We propose Fast-TD-MPC, a lightweight framework that adaptively routes between fast policy execution and test-time planning, reserving costly deliberation for states where it is most needed. Fast-TD-MPC delivers competitive task performance across 103 continuous control tasks while achieving up to ~4x faster inference. Under external disturbances, Fast-TD-MPC selectively falls back to planning, maintaining robustness comparable to the original planner.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Latent dynamics models and model-based RL, one of 11 directions in World models. Builds on 58 of its 72 references that are in the map. First posted on arXiv 26 Sep 2026.

Builds on

58 of 72 references.

Closest ancestors · 7

Same direction, newest first

  1. 2025TD-MPC-Opt: Distilling Model-Based Multi-Task Reinforcement Learning Agents2 Jul
  2. 2025Discrete Codebook World Models for Continuous Control1 Mar
  3. 2023TD-MPC2: Scalable, Robust World Models for Continuous Control25 Oct
  4. 2023Mastering Diverse Domains through World Models10 Jan
  5. 2019Dream to Control: Learning Behaviors by Latent Imagination3 Dec
  6. 2018Recurrent World Models Facilitate Policy Evolution4 Sep
  7. 2017Neural Network Dynamics for Model-Based Deep Reinforcement Learning with Model-Free Fine-Tuning8 Aug

Other references · 51

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  2. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  3. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
  4. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  5. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  6. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  7. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  8. 2023Accelerating Large Language Model Decoding with Speculative Samplingcited by 1,006
  9. 2020SAPIEN: A SimulAted Part-based Interactive ENvironmentcited by 595
  10. 2016Adaptive Computation Time for Recurrent Neural Networkscited by 506
  11. 2022Temporal Difference Learning for Model Predictive Controlcited by 397
  12. 2023ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skillscited by 327
38 more
  1. 2015Model Predictive Path Integral Control using Covariance Variable Importance Samplingcited by 124
  2. 2022MyoSuite -- A contact-rich simulation suite for musculoskeletal motor controlcited by 80
  3. 2024World Models for Autonomous Driving: An Initial Surveycited by 80
  4. 2022Simplicial Embeddings in Self-Supervised Learning and Downstream Classificationcited by 24
  5. 2025Bootstrapped Model Predictive Controlcited by 16
  6. 2026Latent Geometry Beyond Search: Amortizing Planning in World Modelscited by 15
  7. 2026Fast LeWorldModelcited by 15
  8. 2023IQL-TD-MPC: Implicit Q-Learning for Hierarchical Model Predictive Controlcited by 14
  9. 2025TD-M(PC)2: Improving Temporal Difference MPC Through Policy Constraintcited by 14
  10. 2025Robust Gymnasium: A Unified Modular Benchmark for Robust Reinforcement Learningcited by 13
  11. 2022Learning Low-Frequency Motion Control for Robust and Dynamic Robot Locomotioncited by 12
  12. 2026VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Modelcited by 9
  13. 2026Hallucination in World Models is Predictable and Preventablecited by 8
  14. 2017Planning, Fast and Slow: A Framework for Adaptive Real-Time Safe Trajectory Planningcited by 7
  15. 2025Guaranteeing Out-Of-Distribution Detection in Deep RL via Transition Estimationcited by 6
  16. 2026Test-Time Scaling for World Action Models via Zero-Shot Geometric Evaluationcited by 6
  17. 2021A Distance-based Anomaly Detection Framework for Deep Reinforcement Learningcited by 5
  18. 2026Open-Loop Planning, Closed-Loop Verification: Speculative Verification for VLAcited by 5
  19. 2025TD-GRPC: Temporal Difference Learning with Group Relative Policy Constraint for Humanoid Locomotioncited by 3
  20. 2025Generative Models From and For Sampling-Based MPC: A Bootstrapped Approach For Adaptive Contact-Rich Manipulationcited by 3
  21. 2025Accelerating Multi-modal LLM Gaming Performance via Input Prediction and Mishit Correctioncited by 3
  22. 2026AEGIS: A Backup Reflex for Physical AIcited by 3
  23. 2026AdaReP:Adaptive Re-Planning under Model Mismatch for Neural World-Model Predictive Controlcited by 3
  24. 2026ELASTIC: Efficiently Learning to Adaptively Scale Test-Time Compute for Generative Control Policiescited by 3
  25. 2026LeFlow: Generative Latent Flow Planning for World Modelscited by 3
  26. 2026EfficientTDMPC: Improved MPC Objectives for Sample-Efficient Continuous Controlcited by 2
  27. 2026TurboMPC: Fast, Scalable, and Differentiable Model Predictive Control on the GPUcited by 2
  28. 2026Finding the Time to Think: Learning Planning Budgets in Real-Time RLcited by 2
  29. 2026Learning When to Stop: Prefix-Optimal Dynamic Diffusion Policies for Continuous Controlcited by 2
  30. 2026AutoIntervene: Calibrated Intervention for Action-Chunking Imitation Learning Policiescited by 2
  31. 2017Event-Triggered Intermittent Sampling for Nonlinear Model Predictive Controlcited by 1
  32. 2024Characterizing and Optimizing Real-Time Optimal Control for Embedded SoCscited by 1
  33. 2025DoublyAware: Dual Planning and Policy Awareness for Temporal Difference Learning in Humanoid Locomotioncited by 1
  34. 2026Adaptive Reinforcement and Model Predictive Control Switching for Safe Human-Robot Cooperative Navigationcited by 1
  35. 2026Dream-MPC: Gradient-Based Model Predictive Control with Latent Imaginationcited by 1
  36. 2026LIDAR-AD: A Decoder-Free Latent-Interaction Dreamer with Action-Residual Chains for Autonomous Drivingcited by 1
  37. 2026OOD-RL-Bench: A Benchmark Framework for Out-of-Distribution Detection in Reinforcement Learningcited by 1
  38. 2026When to Plan: Learning to Select Between Reactive Control and Deliberative Planningcited by 1

50 most cited of 51 All 51 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Bridging Large-Model Reasoning and Real-Time Control via Agentic Fast-Slow Planning
  2. 2024Faster Model Predictive Control via Self-Supervised Initialization Learning
  3. 2025Learning to Plan, Planning to Learn: Adaptive Hierarchical RL-MPC for Sample-Efficient Decision Making
  4. 2026Beyond Policy Alignment: Closing the Planning-Learning Loop for Robot Control with Learned World Models
  5. 2025Model-Based Diffusion Sampling for Predictive Control in Offline Decision Making
  6. 2026Drift is a Sampling Error: SNR-Aware Power Distributions for Long-Horizon Robotic Planning
  7. 2026Solving Markov Decision Processes with Future Information via MPC
  8. 2022Learning to Optimize in Model Predictive Control