mlsanityData to 14 Oct 2026

R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation

arXiv 2603.18202 ↗First posted 18 Mar 2026cs.LG cs.AI cs.ROcited by 10

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

also in 1 more
Abstract (arXiv)

A central challenge in image-based Model-Based Reinforcement Learning (MBRL) is to learn representations that distill essential information from irrelevant visual details. While promising, reconstruction-based methods often waste capacity on large task-irrelevant regions. Decoder-free methods instead learn robust representations by leveraging Data Augmentation (DA), but reliance on such external regularizers limits versatility. We propose R2-Dreamer, a decoder-free MBRL framework with a self-supervised objective that serves as an internal regularizer, preventing representation collapse without resorting to DA. The core of our method is a redundancy-reduction objective inspired by Barlow Twins, which can be easily integrated into existing frameworks. On DeepMind Control Suite and Meta-World, R2-Dreamer is competitive with strong baselines such as DreamerV3 and TD-MPC2 while training 1.59x faster than DreamerV3, and yields substantial gains on DMC-Subtle with tiny task-relevant objects. These results suggest that an effective internal regularizer can enable versatile, high-performance decoder-free MBRL. Code is available at https://github.com/NM512/r2dreamer.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Placed in Latent dynamics models and model-based RL, one of 11 directions in World models. Builds on 35 of its 39 references that are in the map. First posted on arXiv 18 Mar 2026.

Builds on

35 of 39 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025Learning Transformer-based World Models with Contrastive Predictive Coding6 Mar
  2. 2024Efficient World Models with Context-Aware Tokenization27 Jun
  3. 2024Diffusion for World Modeling: Visual Details Matter in Atari20 May
  4. 2024EfficientZero V2: Mastering Discrete and Continuous Control with Limited Data1 Mar
  5. 2023TD-MPC2: Scalable, Robust World Models for Continuous Control25 Oct
  6. 2023STORM: Efficient Stochastic Transformer based World Models for Reinforcement Learning14 Oct
  7. 2022Transformers are Sample-Efficient World Models1 Sep
  8. 2022DreamingV2: Reinforcement Learning with Discrete World Models without Reconstruction1 Mar
  9. 2021Mastering Atari Games with Limited Data30 Oct
  10. 2021DreamerPro: Reconstruction-Free Model-Based Reinforcement Learning with Prototypical Representations27 Oct
  11. 2020Mastering Atari with Discrete World Models5 Oct
  12. 2019Dream to Control: Learning Behaviors by Latent Imagination3 Dec

Other references · 23

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  3. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  4. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  5. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  6. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  7. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  8. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  9. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  10. 2021VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learningcited by 1,057
  11. 2018DeepMind Control Suitecited by 956
  12. 2016Deep Variational Information Bottleneckcited by 930
11 more
  1. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  2. 2020Reinforcement Learning with Augmented Datacited by 468
  3. 2022Temporal Difference Learning for Model Predictive Controlcited by 397
  4. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 372
  5. 2022Masked World Models for Visual Controlcited by 170
  6. 2017Visualizing and Understanding Atari Agentscited by 141
  7. 2021The Distracting Control Suite -- A Challenging Benchmark for Reinforcement Learning from Pixelscited by 100
  8. 2021Temporal Predictive Coding For Model-Based Planning In Latent Spacecited by 55
  9. 2021Unsupervised Visual Attention and Invariance for Reinforcement Learningcited by 30
  10. 2022A Comprehensive Survey of Data Augmentation in Visual Reinforcement Learningcited by 25
  11. 2020Predictive Coding for Locally-Linear Controlcited by 17

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 10.

10 papers.

  1. 2026PH-Dreamer: A Physics-Driven World Model via Port-Hamiltonian Generative Dynamicscited by 3
  2. 2026ARB4WM: An Adversarial Robustness Benchmark for World Models in Continuous Controlcited by 2
  3. 2026DREAM-Chunk: Reactive Action Chunking with Latent World Modelcited by 2
  4. 2026Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failurecited by 1
  5. 2026LIDAR-AD: A Decoder-Free Latent-Interaction Dreamer with Action-Residual Chains for Autonomous Drivingcited by 1
  6. 2026TaskSense: Focusing on What Matters in World Modelscited by 0
  7. 2026TrojanWorld: Backdooring World-Model Agents via Imagination Steeringcited by 0
  8. 2026StarWM: Self-Supervised Trained Attention Routing for Robust World Modelscited by 0
  9. 2026I Act Therefore I Am: When Is JEPA's Action-Conditioning Enough to Learn Causal Mechanisms?cited by 0
  10. 2026Lucid Dreaming for World Models: Learning to Doubt Imagination and Decide by Trustcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2020Dreaming: Model-based Reinforcement Learning by Latent Imagination without Reconstruction
  2. 2026R2R2: Robust Representation for Intensive Experience Reuse via Redundancy Reduction in Self-Predictive Learning
  3. 2026Dreamer-CDP: Improving Reconstruction-free World Models Via Continuous Deterministic Representation Prediction
  4. 2025Dream to Generalize: Zero-Shot Model-Based Reinforcement Learning for Unseen Visual Distractions
  5. 2026VIGOR: Zero-Shot Visual Generalization via Latent-Space Consistency in Model-Based Reinforcement Learning
  6. 2026Reinformed Dreamer: An Asymmetric World Model Efficiently Trained through Latent Guidance
  7. 2024The Surprising Ineffectiveness of Pre-Trained Visual Representations for Model-Based Reinforcement Learning
  8. 2024MuDreamer: Learning Predictive World Models without Reconstruction