Causal World Models

Follow this topic Ask about this topic

Causal world models aim to represent not only what a scene looks like, but how entities and environments change under actions, so agents can plan and answer what-if questions. Methods range from structural causal graphs and object-centric latent dynamics to video-prediction models trained with paired interventions, counterfactual rollouts, or physical constraints. They are tested in simulated physics and manipulation, driving and navigation, and video benchmarks that compare factual futures with altered-action outcomes. Models increasingly improve action control and intervention prediction, but plausible video is not necessarily causally correct: learning effects from limited interventions while retaining factual accuracy and generalizing across settings remains difficult.

Latest papers

47 papers8 papers in the last four weeks, down 27% on the four weeks before. 0.0% of all new papers.

Outcome-Aware Agents: When Does Causal Structure Help a Forward Model Foresee the Consequences of Actions?
Oct 25, 2026Mallesh MadapathiCausal World Models Action-Conditioned World Models
ThinkingDBx Pvt. Ltd., Hyderabad, India
Causal world models under physical parameter shifts
Oct 21, 2026Chu Lin, Yi SunCausal World Models World Models for Robotics
Department of Computer Science, The City College of New York, New York, NY 10031, USA · Department of Electrical Engineering, The City College of New York, New York, NY 10031, USA
CaMeWM: A Causal Medical World Model for Generative Tumor Evolution and Treatment Planning
Oct 17, 2026Huanlong Gao, Xuelei He, Fengjun Zhao +1Causal World Models Causal Counterfactual Generation
School of Electronic Information, Northwest University, No. 1 Xuefu Avenue, Xi’an, 710127, China · Xi’an Key Lab of Radiomics and Intelligent Perception, Northwest University, No. 1 Xuefu Avenue, Xi’an, 710127, China
A Plausible Future Is Not a Causal Future: Can Video Models Generate How the World Unfolds?
Oct 11, 2026Jialiang Li, Zhengning Xu, Zongyang Ma +1Causal Video Generation Video Generation
DeepDelta Lab, School of Artificial Intelligence, Shanghai Jiao Tong University · Xi’an Jiaotong-Liverpool University
CAG-WAM: Causal Action-Gain World Action Models for Robot Manipulation
Oct 10, 2026Peizheng Guo, Cheng Zhang, Xingyu Zhang +2Action-Conditioned World Models Robotic Manipulation
National Key Laboratory of Space Integrated Information System, Institute of Software Chinese Academy of Sciences, Beijing, China · University of Chinese Academy of Sciences, Beijing, China · Roban
CausalDreamer: Learning Predictive World Models with Latent Disentanglement
Oct 8, 2026Prince Jha, Nils Lukas, Kun Zhang +1World Model Learning Causal World Models
Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) · Carnegie Mellon University
Same Predictions, Different Harms: Causal Auditing of Patient World Models
Oct 4, 2026Yicheng Qi, Xiyi XiongStructural Causal Models Causal World Models
Imperial College London, London, United Kingdom
Do-JEPA: From Masking to Intervention in Latent World Models
Sep 29, 2026Hossein Resani, Javen Qinfeng ShiCausal Effect Estimation Causal World Models
Australian Institute for Machine Learning, Adelaide University · Responsible AI Research Centre
World Agent: Can Language Models Keep a World Running?
Sep 26, 2026Weixing Chen, Weipeng Zhang, Nan An +2World Model Evaluation World Models
Sun Yat-sen University · South China University of Technology · Guangdong Key Laboratory of Big Data Analysis and Processing +2
OneWorld: Learning Consistent Physics Across Actions in World Models
Sep 25, 2026Ke He, Yichen Ding, Bin YangAction-Conditioned World Models Physical Consistency in Video Generation
Wuhan University Wuhan, China
Stable and Counterfactually Robust Physical World Models from Imposed Structure and Learned Physics
Sep 24, 2026Yufeng Wang, Parivesh Priye, Lu Wei +1World Model Learning Physics-Informed ML
Stony Brook University · Georgia Institute of Technology · Westlake University
What Matters in Designing World Action Models: An Empirical Study
Sep 21, 2026Chao Tang, Haoqing Wang, Zilang Cen +9World Model Learning Causal World Models
Samsung Robotics eXperience · Samsung R&D Institute China–Beijing · Zhongguancun Academy +2
OnlineWM: Causality-Aware Active Online Learning for Effective World Modeling
Sep 20, 2026Yikun Miao, Fangqi Zhu, Quanxin Shou +6World Model Learning Causal World Models
Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong SAR, China
CausalWM: Causal Chain-of-Thought Reasoning for Embodied World Model
Sep 19, 2026Ziming Xu, Shuang Liang, Ruobing Han +10Causal World Models Action-Conditioned World Models
Aether AI · University of California, San Diego · Vanderbilt University
Compact but Moving: Intervention-Relevant Geometry in Recurrent World Models
Sep 18, 2026Yuming Chen, Yang LiuWorld Models World Model Learning
College of Intelligent Robotics and Advanced Manufacturing, Fudan University Shanghai, China
Converting Sequenced Fuzzy Cognitive Maps to Causal Virtual Worlds with Large Video Generators
Sep 14, 2026Akash Kumar Panda, Olaoluwa Adigun, Bart KoskoCausal World Models Fuzzy Cognitive Maps
Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA · School of Computing and Information Sciences, Florida International University Miami, FL 33199, USA
InfluenceField: A Differentiable Field with Interventionally Identifiable Causal Structure for Multimodal World Modeling
Sep 7, 2026Zihao Yang, Zijia Wang, Zhiqiu HuangCausal Representation Learning Causal World Models
Mathematical Institute, University of Oxford · Department of Computer Science, University of Oxford · School of Mathematical Sciences, University of Nottingham Ningbo China
CST-WM: A Causally Structured World Model for Embodied Visual Tracking
Sep 5, 2026Junyi Hu, Shuaihang Yuan, Jiazhao Liang +1Causal World Models Visual Object Tracking
New York University Abu Dhabi
CAER: Causal Action Effect Reweighting for World Model Training
Aug 31, 2026Jianjie Fang, Xvyuan Liu, Ziyou Wang +9Action-Conditioned Video Generation World Model Learning
1Tsinghua University · University of Science and Technology of China · 2Manifold AI
A Unifying Perspective on Causal World Models: From Observations to Representations to Structure
Aug 13, 2026Avinash Kori, Fabrizio RussoCausal Representation Learning World Model Learning
Department of Computing, Imperial College London, London, UK
FACT: Failure-Aware Causal Training for World-Action Models
Aug 10, 2026Quanquan Peng, Yutong Liang, Rui Yan +2Causal World Models Action-Conditioned World Models
University of California San Diego
Compiling VGDL into Causal Models
Aug 10, 2026Mohit Jiwatode, Bodo Rosenhahn, Alexander DockhornCausal World Models Structural Causal Models
Institute for Information Processing Leibniz University Hannover Hannover, Germany · SDU Metaverse Lab University of Southern Denmark Odense, Denmark
Twin Rollouts: Noise-Coupled Counterfactual Branching in Interactive Video World Models
Aug 10, 2026Yu Ma, Hongli Shi, Xinran XuAction-Conditioned Video Generation Counterfactual Evaluation
CausalNav: Reliability-Certified Causal World Models for Control under Physical-Parameter Shift
Aug 7, 2026Yiyao Zhang, Diksha Goel, Hussain AhmadCausal World Models World Models
School of Computing and Information Technology, University of Wollongong, Wollongong, NSW, Australia · CSIRO’s Data61, Clayton, VIC, Australia · School of Computer Science and Information Technology, Adelaide University, Adelaide, SA, Australia
Learning Implicit Causal World Models from Multi-Agent Demonstrations
Jul 28, 2026Jasorsi GhoshCausal Representation Learning Causal World Models
Department of Computer Science Purdue University West Lafayette, Indiana, USA
A cubical formalisation of topos causal models: intervention, forcing, and a contextuality obstruction
Jul 17, 2026Karen SargsyanCausal World Models Causal Intervention
Institute of Chemistry, Academia Sinica, Taipei, Taiwan
When Do Causal World Models Help Modular LLM Agents
Jul 12, 2026Xinyuan Song, Zekun CaiCausal World Models LLM Agents
Emory University, Atlanta, GA, USA · The University of Tokyo, Tokyo, Japan · LocationMind, Tokyo, Japan
OWMDrive: Causality-Aware End-to-End Autonomous Driving via 4D Occupancy World Model
Jun 29, 2026Junjie Cheng, Ruiqi Song, Ye Wu +3Causal World Models End-to-End Autonomous Driving
The School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China · Waytous Inc., Qingdao 266109, China · The State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China +1
Causal Reward World Models: Zero-shot Reward Design for Automated Skill Generation
Jun 22, 2026Yang Yang, Yuchuang Tong, Zhengtao Zhang +6RL for Robotics Robot Skill Learning
Institute of Automation, Chinese Academy of Sciences, Beijing 100049, China · Intelligent Manufacturing Institute, HFUT, Hefei 230051, China · School of Electrical Engineering and Automation, Anhui University, Hefei 230061, China +1
Self-Evolving Cognitive Framework via Causal World Modeling for Embodied Scientific Intelligence
Jun 21, 2026Yi Yu, Tetsunari InamuraCausal World Models Embodied AI
Graduate School of Advanced Science and Engineering, Hiroshima University, Japan · Advanced Intelligence and Robotics Research Center, Brain Science Institute, Tamagawa University, Japan

Oct 25, 2026J/K move · Enter open · S save

Outcome-Aware Agents: When Does Causal Structure Help a Forward Model Foresee the Consequences of Actions?

Mallesh Madapathi

ThinkingDBx Pvt. Ltd., Hyderabad, India

Causal World Models Action-Conditioned World Models Model-Based Planning Causal Structure Learning

Open paper AskPDF arXiv

Agents that act on real systems should predict what their actions will cause before acting. We study when causal structure helps a learned forward model do this, using a planner that rejects candidate actions whose simulated multi-step outcome is unsafe, in synthetic structural causal models where the true consequences are known. We compare the Action-Conditioned Latent Structural Causal Model (AC-LSCM: per-factor mechanisms over a learned sparse graph, actions as do-interventions) with three unstructured forward models on seven random graphs with 10 or 20 variables, a task configured by a fixed rule, and two controls: a do-wrapper that gives every model the same action semantics, and counterfactual data given to all models or to none. When its factors are the observed variables, AC-LSCM recovers the true graph exactly on every 10-variable graph and has lower three-step prediction error than every baseline in all 396 paired comparisons. Its excess safety violations over an oracle are 0.25 of the no-model level on 10-variable graphs (best baseline 0.31) and 0.41 on 20-variable graphs (0.53), and its margin is largest for two-variable actions never seen in training (0.18 versus 0.28). With learned latent factors the same model is worse than the baselines; the choice of factors matters more than the explicit do-operator or counterfactual data, and every planner stays far from the oracle. This version corrects an earlier one, whose model learned an empty graph and whose headline 36x safety improvement came from a task that tested only extrapolation of action values; we withdraw that claim. Code and all per-seed results are released.