A self-correcting multi-agent LLM framework for language-based physics simulation and explanation

Physics-based simulations are essential in science and engineering, yet creating them typically requires expert knowledge of numerical solvers and governing equations. Large language models (LLMs) offer new possibilities for natural language-based simulation, but they often fail when prompts are vague, incomplete, or multilingual. We present MCP-SIM (Memory-Coordinated Physics-Aware Simulation), a self-correcting multi-agent framework that transforms underspecified prompts into validated simulations and explanatory reports. The system integrates input clarification, code generation, error diagnosis, and multilingual explanation through structured agent collaboration and persistent memory. Rather than relying on one-shot code generation, MCP-SIM emulates expert-like reasoning via iterative plan–act–reflect–revise cycles. Across twelve tasks of increasing complexity, the framework solved all benchmark cases and improved convergence efficiency relative to GPT-based and human-in-the-loop baselines, under the specific metrics defined in this study. In addition to numerical accuracy, the system produces interpretable, language-localized reports that explain each simulation’s physical logic. MCP-SIM represents a step toward general-purpose autonomous scientific assistants that simulate, adapt, and teach through natural language. While these results indicate strong robustness on the tested suite, performance in specialized domains and under distributions beyond our benchmark remains an area for future validation.