Abductive World Modeling via Causal Representation Learning

Ziqi Liu; Songhan Yang; Linfan Zhou; Jiatong Liu; Lijun Peng; Long Wan; Yinqi Bai

cs.LGSep 29, 2026

Abductive World Modeling via Causal Representation Learning

Authors: Ziqi Liu, Songhan Yang, Linfan Zhou, Jiatong Liu, Lijun Peng, Long Wan, Yinqi Bai

Organizations: BAIZ Team

Abstract

The central challenge of world modeling is to learn representations that capture how the world evolves. However, existing world models predominantly represent future states without explicitly capturing the latent causes underlying their evolution, limiting their ability to reason about why and how the world changes. To address this limitation, we propose Abductive World Modeling (AWM), a framework that learns structured causal representations by abductively inferring latent causes from predicted futures. Specifically, we realize AWM through the Hierarchical Abductive State Pyramid (HASP), which organizes the inferred world state into three complementary components - Entity, Dynamic, and Relation - capturing what exists, how it changes, and how entities interact, respectively. By jointly reasoning over the current observation and its predicted future, HASP abductively infers these latent factors and integrates them into a structured state representation for downstream reasoning. To the best of our knowledge, AWM is the first framework to introduce abductive state inference into latent-space world modeling for learning structured representations of world dynamics. Experiments across physical prediction, causal reasoning, and action understanding demonstrate the effectiveness of our approach. Compared with V-JEPA, a state-of-the-art latent-space world model, AWM improves physical prediction AUROC by 10.7%, causal reasoning accuracy by 16.8%, and action Top-1 accuracy by 68.0%.

Figures & tables

Figure 1 : Overview of the AWM. Given an observed context, a frozen predictive visual backbone first predicts its future latent state. The HASP reasons over the current and predicted future representations to infer Entity, Dynamic, and Relation states for downstream prediction and reasoning.

	Physion++			CLEVRER		EK100 (T1/T5)
Method	AUROC	Bal. Acc.	Acc.	Option Acc.	Question Acc.	Verb	Noun	Action
V-JEPA 2	$65.20$	$61.57$	$61.60$	$65.46$	$42.20$	$55.74/86.81$	$34.65/62.67$	$25.67/46.94$
Orca	$59.15$	$55.94$	$56.64$	—	—	$33.75/74.39$	$21.54/45.36$	$13.08/30.46$
VideoMAE v2	$67.61$	$62.60$	$62.86$	—	—	$59.54/82.57$	$46.37/68.79$	$35.96/54.57$
AWM (Ours)	$\mathbf{72.19}$	$\mathbf{66.17}$	$\mathbf{65.63}$	$\mathbf{70.89}$	$\mathbf{49.31}$	$\mathbf{69.05/91.27}$	$\mathbf{47.59/75.05}$	$\mathbf{43.12/67.34}$

Table 1 : Main comparison across datasets. All entries are percentage values reported without the percent sign, and the best value in each metric column is boldfaced. “Bal. Acc.” and “Acc.” denote balanced accuracy and accuracy, respectively, while EK100 results are reported as Top-1/Top-5. Orca and VideoMAE v2 are marked as “—” on CLEVRER because their released models do not provide the predictive module required by the CLEVRER evaluation pipeline.

Panel of Figure 2 — Figure 2 : Qualitative visualization of Entity grounding on EK100 and CLEVRER. (a)–(c) show representative EK100 scenes, where predicted Entity regions align with ground-truth hands and manipulated objects. (d) shows consecutive CLEVRER frames, where matched Entity predictions remain aligned with the corresponding ground-truth objects over time. Green boxes denote ground-truth regions and magenta boxes denote matched predicted regions.

State	Factor	Metric	Score
Entity	Presence	AUROC	0.9634
	Spatial extent	$R^{2}$	0.5316
Dynamic	Speed	$R^{2}$	0.7198
	Signed velocity	$R^{2}$	0.3805
Relation	Contact	AUROC	0.9789
	Pairwise distance	$R^{2}$	0.5815

Table 2 : Native factor readability on Physion++.

Figure 3 : Interaction prediction with individual HASP states on CLEVRER. Full uses all three states.

Figure 4 : Entity intervention on Physion++ OCP prediction. Target masks the target Entity slot, Irrelevant masks a non-target slot, and Full retains all Entity slots. Performance is evaluated by accuracy, balanced accuracy, and AUROC (

\uparrow

), showing how prediction quality changes when target-relevant or irrelevant entity information is removed.

Condition	Speed MAE ( $\downarrow$ )	Contact AUROC ( $\uparrow$ )	OCP AUROC ( $\uparrow$ )
AWM	0.0670	0.9837	0.7907
w/o Temporal	0.1120	0.9485	0.5646

Table 3 : Temporal-static latent intervention on Physion++. AWM uses the original context and future features. AWM w/o Temporal temporally averages and repeats these features at each time step, removing temporal variation while keeping the model and readouts frozen.

Condition	Pair-dist. MAE ( $\downarrow$ )	TTC MAE ( $\downarrow$ )	Contact AUROC ( $\uparrow$ )
AWM	0.0196	0.1306	0.9633
w/o Contact Pair	0.0225	0.1376	0.9488
w/o Non-contact Pair	0.0208	0.1349	0.9518

Table 4 : Relation-pair intervention on CLEVRER. AWM uses all Relation pairs; AWM w/o Contact Pair removes the target contact pair; and AWM w/o Non-contact Pair removes an unrelated pair. We report geometric and interaction prediction performance.

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Diagnostic	Value
Presence F1	99.77
Slot occupancy	64.52
Identity consistency	70.29
ID switch rate	4.05
Target-object accuracy	95.79
Object-type accuracy	92.86

Appendix

Table 5 : Additional diagnostics of the Physion++ Entity slots.

Condition	TP	TN	FP	FN
Visual-only	237	292	129	142
Visual + ShallowProbe	277	297	124	102
Target mask	325	181	240	54
Irrelevant mask	262	305	116	117
Random slot	279	294	127	100

Appendix

Table 6 : Physion++ OCP confusion counts under probe interventions.

Condition	Mean change	Mean absolute change
Target mask	+0.49195	0.67680
Irrelevant mask	-0.08051	0.17355
Random slot	+0.02437	0.08895

Appendix

Table 7 : Paired OCP-logit changes under probe-level interventions.

Condition	Question Correct	Question Acc. $\uparrow$	Option Correct	Option Acc. $\uparrow$
AWM	1,785/3,557	50.18%	5,099/7,114	71.68%
AWM w/o Relevant Object	1,546/3,557	43.46%	4,752/7,114	66.80%
AWM w/o Irrelevant Object	1,684/3,557	47.34%	4,948/7,114	69.55%

Appendix

Table 8 : Relevant-object intervention on CLEVRER predictive queries. AWM uses all Entity evidence; AWM w/o Relevant Object removes the object referred to by the query; and AWM w/o Irrelevant Object removes an object unrelated to the query.

Parameter	Value
Predictive Backbone
Backbone	V-JEPA 2 ViT-H
Observed frames	16
Predicted future frames	16
Patch size	16
Tubelet size	2

Appendix

Table 9 : Main configuration and hyperparameter settings of AWM. We report the predictive backbone, evaluation configuration, and computational setup used in our experiments.

Explore similar work

CausalWM: Causal Chain-of-Thought Reasoning for Embodied World Model
Sep 19, 2026Ziming Xu, Shuang Liang, Ruobing Han +10Efficient World-Action Model Future Video Prediction
Aether AI · University of California, San Diego · Vanderbilt University
A Unifying Perspective on Causal World Models: From Observations to Representations to Structure
Aug 13, 2026Avinash Kori, Fabrizio RussoCausal Representation Learning World Models
Department of Computing, Imperial College London, London, UK
Latent State Design for World Models under Sufficiency Constraints
May 3, 2026Keon Woo KimLatent World Models World Models
Om Labs

Sep 19, 2026, cs.CVJ/K move · Enter open · S save

CausalWM: Causal Chain-of-Thought Reasoning for Embodied World Model

Ziming Xu, Shuang Liang, Ruobing Han, Ziqiao Xi+9

Aether AI · University of California, San Diego · Vanderbilt University

Efficient World-Action Model Future Video Prediction Causal Chain-of-Thought Reasoning Reinforcement Learning Post-Training

Open paper AskPDF arXiv

Embodied world models learn to predict future physical dynamics from visual observations and control signals, where physical knowledge is implicitly entangled within latent representations. We introduce CausalWM, a 16B embodied world model that performs explicit causal chain-of-thought reasoning before future video prediction. CausalWM organizes useful variables into a reasoning trajectory, allowing the model to progressively capture causal dependencies underlying physical evolution. To train CausalWM, we collect 31K hours embodied data and develop a three-stage paradigm consisting of large-scale video pre-training, causal CoT mid-training, and multi-objective RL post-training. Despite using only a limited set of supervised CoT variables, CausalWM exhibits emergent in-context learning capabilities, enabling contextual visual feature guidance and efficient few-step generation. CausalWM achieves state-of-the-art performance across language-conditioned, action-conditioned, single-view and multi-view benchmarks, including Top-1 performance on TriWorldBench leaderboard.

Abstract

Explore similar work

CausalWM: Causal Chain-of-Thought Reasoning for Embodied World Model

A Unifying Perspective on Causal World Models: From Observations to Representations to Structure

Latent State Design for World Models under Sufficiency Constraints