mlsanityData to 14 Oct 2026

BRICKS-WM: Building Reusability via Interface Composition Kinetics for Structured World Models

arXiv 2606.16489 ↗First posted 15 Jun 2026cs.LGcited by 0

On the map: World models → Latent predictive world models

also in 1 more
Abstract (arXiv)

Model-based Reinforcement Learning (MBRL) has achieved remarkable success in continuous control by leveraging latent world models. However, prevailing approaches typically rely on monolithic latent dynamics, entangling environment dynamics into a coupled process. This coupling severely limits reusability: altering the agent necessitates retraining the entire world from scratch, even if the environment remains constant. To address this, we introduce BRICKS-WM (Building Reusability via Interface Composition Kinetics for Structured World Models), a framework for the modular assembly of structured world models. Driven by the insight that the physical world is composed of independent entities, we posit that global dynamics can be modeled as a composition of distinct dynamical modules interacting via latent interfaces. As a minimal instantiation, we factorize the latent state space into an actuated Agent module and an external Background module, bridged by a learned latent interface. Unlike prior object-centric methods that prioritize visual segmentation, BRICKS-WM enforces a functional separation in transition dynamics, ensuring that background dynamics remains agnostic to the agent's dynamics. Empirically, BRICKS-WM achieves control performance comparable to strong monolithic baselines when trained from scratch, and enables the reuse of frozen background dynamics across agents.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Latent predictive world models, one of 11 directions in World models. Builds on 39 of its 41 references that are in the map. First posted on arXiv 15 Jun 2026.

Builds on

39 of 41 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Learning Interactive World Model for Object-Centric Reinforcement Learning4 Nov
  2. 2025Dyn-O: Building Structured World Models with Object-Centric Representations4 Jul
  3. 2022Temporal Difference Learning for Model Predictive Control9 Mar
  4. 2021Conditional Object-Centric Learning from Video24 Nov
  5. 2019Contrastive Learning of Structured World Models27 Nov
  6. 2018Learning Latent Dynamics for Planning from Pixels12 Nov

Other references · 33

  1. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  2. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  3. 2023Mastering Diverse Domains through World Modelscited by 1,253
  4. 2020Mastering Atari with Discrete World Modelscited by 1,007
  5. 2018DeepMind Control Suitecited by 956
  6. 2020Object-Centric Learning with Slot Attentioncited by 876
  7. 2024Genie: Generative Interactive Environmentscited by 747
  8. 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 512
  9. 2016Interaction Networks for Learning about Objects, Relations and Physicscited by 509
  10. 2019Model-Based Reinforcement Learning for Ataricited by 502
  11. 2019MONet: Unsupervised Scene Decomposition and Representationcited by 392
  12. 2019Multi-Object Representation Learning with Iterative Variational Inferencecited by 360
21 more
  1. 2022Transformers are Sample-Efficient World Modelscited by 328
  2. 2024Diffusion for World Modeling: Visual Details Matter in Ataricited by 300
  3. 2019GENESIS: Generative Scene Inference and Sampling with Object-Centric Latent Representationscited by 227
  4. 2019Recurrent Independent Mechanismscited by 219
  5. 2020SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decompositioncited by 190
  6. 2021Embodied Intelligence via Learning and Evolutioncited by 163
  7. 2016Learning Modular Neural Network Policies for Multi-Task and Multi-Robot Transfercited by 127
  8. 2020One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Controlcited by 121
  9. 2019Entity Abstraction in Visual Model-Based Reinforcement Learningcited by 110
  10. 2022MetaMorph: Learning Universal Controllers with Transformerscited by 89
  11. 2023STORM: Efficient Stochastic Transformer based World Models for Reinforcement Learningcited by 89
  12. 2022Denoised MDPs: Learning World Models Better Than the World Itselfcited by 68
  13. 2021Learning Task Informed Abstractionscited by 58
  14. 2022Iso-Dream: Isolating and Leveraging Noncontrollable Visual Dynamics in World Modelscited by 53
  15. 2021Conditional Neural Relational Inference for Interacting Systemscited by 28
  16. 2024SOLD: Slot Object-Centric Latent Dynamics Models for Relational Manipulation Learning from Pixelscited by 24
  17. 2024Slot State Space Modelscited by 17
  18. 2025Scaling Cross-Embodiment World Models for Dexterous Manipulationcited by 13
  19. 2023Efficient RL via Disentangled Environment and Agent Representationscited by 12
  20. 2024DEAR: Disentangled Environment and Agent Representations for Reinforcement Learning without Reconstructioncited by 6
  21. 2024Compete and Compose: Learning Independent Mechanisms for Modular World Modelscited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Modeling Unseen Environments with Language-guided Composable Causal Components in Reinforcement Learning
  2. 2026Physically Native World Models: A Hamiltonian Perspective on Generative World Modeling
  3. 2026WALL-SS: Scaling Long-horizon World Models via Next-Scale Autoregression
  4. 2026AcrossWAM1.0:A Modular Latent World-Action Stack for Compact Robot Policies
  5. 2025World-in-World: World Models in a Closed-Loop World
  6. 2026Benchmarking World Models for Continual Learning on Compositional Tasks
  7. 2026DWM: Separating World Effects from Actions in Latent World Models
  8. 2026Enhancing Policy Learning with World-Action Model