Aligning Agentic World Models via Knowledgeable Experience Learning

arXiv 26.1 2026 39.6 method, system

TLDR

WorldMind aligns LLM agents with physical laws by constructing a symbolic knowledge repository from environmental feedback, reducing physical hallucinations.

Reasoning

The paper introduces a novel framework that uses symbolic knowledge to address physical hallucinations without heavy retraining, showing strong cross-model transfer. However, experiments are limited to simulated benchmarks, and scalability to complex real-world dynamics remains unclear.

Read-first score

Read-first score 39.6, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 45.

Recency 5%

100

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

Methodology quality 15%

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

Topical relevance 25%

64.3

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

Reproducibility 15%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=repository

Graph prestige 15%

5.7

PageRank-style prestige over the local citation graph built from available OpenAlex references. local_pagerank=0.0016

Citation impact 15%

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. cited_by_count=0

Citation velocity 10%

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

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 521.

Keyword Scores

world model

world dynamics prediction

interactive world model

world simulator

model-based reinforcement learning world model

generative world model

video world model

Deep Analysis

Innovations

WorldMind framework that autonomously constructs a symbolic World Knowledge Repository by synthesizing environmental feedback
Unification of Process Experience (enforcing physical feasibility via prediction errors) and Goal Experience (guiding task optimality through successful trajectories) for alignment
Demonstrated cross-model and cross-environment transferability without continuous retraining

Methodology

WorldMind constructs a symbolic World Knowledge Repository by synthesizing environmental feedback. It unifies Process Experience, which enforces physical feasibility via prediction errors, and Goal Experience, which guides task optimality through successful trajectories. The framework is evaluated on EB-ALFRED and EB-Habitat benchmarks against baselines.

Key Results

WorldMind achieves superior performance compared to baselines on EB-ALFRED and EB-Habitat, with remarkable cross-model and cross-environment transferability.

Resources

Paper