Score: 2

Benchmark for Planning and Control with Large Language Model Agents: Blocksworld with Model Context Protocol

Published: December 3, 2025 | arXiv ID: 2512.03955v1

By: Niklas Jobs , Luis Miguel Vieira da Silva , Jayanth Somashekaraiah and more

BigTech Affiliations: Siemens

Potential Business Impact:

Tests how smart robots learn new jobs.

Business Areas:
Natural Language Processing Artificial Intelligence, Data and Analytics, Software

Industrial automation increasingly requires flexible control strategies that can adapt to changing tasks and environments. Agents based on Large Language Models (LLMs) offer potential for such adaptive planning and execution but lack standardized benchmarks for systematic comparison. We introduce a benchmark with an executable simulation environment representing the Blocksworld problem providing five complexity categories. By integrating the Model Context Protocol (MCP) as a standardized tool interface, diverse agent architectures can be connected to and evaluated against the benchmark without implementation-specific modifications. A single-agent implementation demonstrates the benchmark's applicability, establishing quantitative metrics for comparison of LLM-based planning and execution approaches.

Country of Origin
🇩🇪 Germany

Repos / Data Links

Page Count
6 pages

Category
Computer Science:
Artificial Intelligence