AGWM: Affordance-Grounded World Models for Environments with Compositional Prerequisites

arXiv 2026 55.5 method

TLDR

AGWM learns a DAG of prerequisite dependencies to track action executability, improving multi-step prediction in environments with compositional prerequisites.

Reasoning

The paper clearly identifies a limitation of standard world models (ignoring action preconditions) and proposes a structured solution (affordance-grounded DAG). Strengths include addressing a specific problem and showing empirical gains in simulated environments. Weaknesses are the lack of real-world experiments and reliance on game-based simulations only.

Read-first score

Read-first score 55.5, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 49.

Recency 6%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Citation impact 18%

74.7

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.74735802

Topical relevance 29%

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Methodology quality 18%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=experiment,result

Reproducibility 18%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Citation velocity 12%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

FrontierBridge

Rank sensitivity

Stability: volatile; rank range: 383.

Keyword Scores

world model

world dynamics prediction

model-based reinforcement learning world model

interactive world model

generative world model

world simulator

video world model

Deep Analysis

Innovations

Explicit modeling of dynamic action executability via a DAG of prerequisite dependencies
Handling structure-changing events (SC events) where actions become executable or non-executable based on prerequisites
Affordance-grounded world model that tracks affordance state to improve multi-step prediction

Methodology

AGWM learns an abstract affordance structure represented as a directed acyclic graph (DAG) of prerequisite dependencies to explicitly track the dynamic executability of actions. The model is evaluated on game-based simulated environments, comparing multi-step prediction error, generalization to novel configurations, and interpretability against standard world models.

Key Results

AGWM achieves lower multi-step prediction error, better generalization to novel configurations, and improved interpretability compared to conventional world models.

Resources

Paper