VLM-DEWM: Dynamic External World Model for Verifiable and Resilient Vision-Language Planning in Manufacturing

arXiv 26.2 2026 56.7 method, application

TLDR

VLM-DEWM decouples VLM reasoning from world-state management using a persistent external world model, improving state tracking and recovery in manufacturing.

Reasoning

The paper presents a novel cognitive architecture that addresses stateless operation and opaque reasoning in VLM-based planning, with strong empirical results on real robots. However, the approach is domain-specific to manufacturing and lacks discussion of limitations or generalization.

Read-first score

Read-first score 56.7, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 35.

Recency 8%

100

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

Methodology quality 25%

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

Topical relevance 42%

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 25%

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

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 405.

Keyword Scores

world model

world dynamics prediction

world simulator

interactive world model

model-based reinforcement learning world model

generative world model

video world model

Deep Analysis

Innovations

Decoupling VLM reasoning from world-state management via a persistent, queryable Dynamic External World Model (DEWM)
Externalizable Reasoning Trace (ERT) comprising action proposal, world belief, and causal assumption for structured decision validation
Discrepancy analysis between predicted and observed states enabling targeted recovery instead of global replanning

Methodology

VLM-DEWM is a cognitive architecture that separates VLM reasoning from world-state management using a Dynamic External World Model (DEWM). Each VLM decision is structured into an Externalizable Reasoning Trace (ERT) containing action proposal, world belief, and causal assumption, which is validated against DEWM before execution. When failures occur, discrepancy analysis between predicted and observed states triggers targeted recovery. The system is evaluated on multi-station assembly, large-scale facility exploration, and real-robot recovery under induced failures, compared against baseline memory-augmented VLM systems.

Key Results

VLM-DEWM improves state-tracking accuracy from 56% to 93%, increases recovery success rate from below 5% to 95%, and significantly reduces computational overhead through structured memory.

Resources

Paper