mlsanityData to 14 Oct 2026

Lucid Dreaming for World Models: Learning to Doubt Imagination and Decide by Trust

arXiv 2609.37156 ↗First posted 29 Sep 2026cs.LG cs.AIcited by 0

On the map: World models → World models for language agents

Abstract (arXiv)

World models enable agents to learn and plan in imagination, but predictions beyond their experience can become unreliable and mislead decisions. Existing uncertainty estimates derived from predictions can remain overconfident on unfamiliar state-action pairs. We propose the Lucid World Model (LucidWM), which learns doubt from experience and propagates trust through imagination. By integrating Subjective Logic into categorical latent transitions, LucidWM distinguishes predicted outcomes from their evidential support and assigns each transition a degree of doubt. The complement of this doubt defines transition-level trust, which accumulates multiplicatively along imagined trajectories to reweight returns for policy learning and guide action selection. Uncertainty estimation requires no additional parameters or forward passes. Evaluated on four base world models against seventeen uncertainty readouts, LucidWM detects environmental changes and signals uncertainty during action-corrupted rollouts. In a controlled navigation case study, acting on trust reduces the number of steps required to reach the goal from 362 to 190. Fifteen demonstration videos show how LucidWM doubts its dreams and acts on that doubt. Videos are available at https://lucidwm.github.io.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in World models for language agents, one of 11 directions in World models. Builds on 38 of its 42 references that are in the map. First posted on arXiv 29 Sep 2026.

Builds on

38 of 42 references.

References · 38

  1. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  3. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  4. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  5. 2018Exploration by Random Network Distillationcited by 993
  6. 2018Recurrent World Models Facilitate Policy Evolutioncited by 785
  7. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 774
  8. 2019When to Trust Your Model: Model-Based Policy Optimizationcited by 773
  9. 2018Evidential Deep Learning to Quantify Classification Uncertaintycited by 732
  10. 2020MOPO: Model-based Offline Policy Optimizationcited by 636
  11. 2022Out-of-Distribution Detection with Deep Nearest Neighborscited by 600
  12. 2020MOReL : Model-Based Offline Reinforcement Learningcited by 590
26 more
  1. 2018Predictive Uncertainty Estimation via Prior Networkscited by 557
  2. 2022DayDreamer: World Models for Physical Robot Learningcited by 400
  3. 2020Planning to Explore via Self-Supervised World Modelscited by 368
  4. 2021Laplace Redux -- Effortless Bayesian Deep Learningcited by 365
  5. 2019Deep Evidential Regressioncited by 355
  6. 2017Snapshot Ensembles: Train 1, get M for freecited by 350
  7. 2024Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllabilitycited by 347
  8. 2022Transformers are Sample-Efficient World Modelscited by 328
  9. 2018Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problemcited by 323
  10. 2018Model-Ensemble Trust-Region Policy Optimizationcited by 219
  11. 2018Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansioncited by 168
  12. 2022Trusted Multi-View Classification with Dynamic Evidential Fusioncited by 168
  13. 2020Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Countscited by 164
  14. 2018Lipschitz Continuity in Model-based Reinforcement Learningcited by 91
  15. 2022Pitfalls of Epistemic Uncertainty Quantification through Loss Minimisationcited by 66
  16. 2024Reliable Conflictive Multi-View Learningcited by 45
  17. 2020Trust the Model When It Is Confident: Masked Model-based Actor-Criticcited by 36
  18. 2024Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?cited by 35
  19. 2019Calibrated Model-Based Deep Reinforcement Learningcited by 31
  20. 2024Trust the Model Where It Trusts Itself -- Model-Based Actor-Critic with Uncertainty-Aware Rollout Adaptioncited by 18
  21. 2025Object-Centric World Models from Few-Shot Annotations for Sample-Efficient Reinforcement Learningcited by 10
  22. 2026R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentationcited by 10
  23. 2025Exploration-Driven Generative Interactive Environmentscited by 9
  24. 2025medDreamer: Model-Based Reinforcement Learning with Latent Imagination on Complex EHRs for Clinical Decision Supportcited by 9
  25. 2025Accurate and Efficient World Modeling with Masked Latent Transformerscited by 7
  26. 2026Biased Dreams: Limitations to Epistemic Uncertainty Quantification in Latent Dynamics Modelscited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Probabilistic Dreaming for World Models
  2. 2026From Generative Engines to Actionable Simulators: The Imperative of Physical Grounding in World Models
  3. 2026Mind Dreamer: Untethering Imagination via Active Causal Intervention on Latent Manifolds
  4. 2026EvoBelief: Falsifiable World Models for Self-Improving LLM Agents
  5. 2026PRISM: PRior-guided Imagination Sampling in world Models
  6. 2025Uncertainty in Action: Confidence Elicitation in Embodied Agents
  7. 2025Mitigating LLM Hallucination via Behaviorally Calibrated Reinforcement Learning
  8. 2025World-in-World: World Models in a Closed-Loop World