DR. WELL: Dynamic Reasoning and Learning with Symbolic World Model for Embodied LLM-Based Multi-Agent Collaboration

arXiv 25.11 2025 44.2 method, system, application

TLDR

A decentralized neurosymbolic framework using a shared symbolic world model for cooperative multi-agent planning, validated on block-push tasks.

Reasoning

The paper introduces a novel approach combining symbolic planning with a dynamic world model to avoid trajectory-level coordination issues. Strengths include clear methodology and empirical results on block-push tasks; weaknesses include limited real-world validation and lack of comparison to baselines in the abstract.

Read-first score

Read-first score 44.2, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 21.

Recency 8%

86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Methodology quality 25%

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

Reproducibility 25%

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

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

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 357.

Keyword Scores

world model

interactive world model

world dynamics prediction

world simulator

generative world model

video world model

model-based reinforcement learning world model

Deep Analysis

Innovations

Two-phase negotiation protocol for role proposal and consensus commitment
Decentralized neurosymbolic framework using symbolic plans instead of raw trajectories
Shared world model that encodes current state and is updated as agents act, capturing reusable patterns

Methodology

DR. WELL is a decentralized neurosymbolic framework for cooperative multi-agent planning. It uses a two-phase negotiation protocol where agents propose candidate roles with reasoning and then commit to a joint allocation under consensus and environment constraints. After commitment, each agent independently generates and executes a symbolic plan for its role, grounded via a shared world model that is updated as agents act.

Key Results

Experiments on cooperative block-push tasks show that agents adapt across episodes, with the dynamic world model capturing reusable patterns and improving task completion rates and efficiency. The method trades a time overhead for evolving, more efficient collaboration strategies through negotiation and self-refinement.

Limitations

Time overhead from negotiation and self-refinement process

Resources

Paper