Language Models Refine Mechanical Linkage Designs Through Symbolic Reflection and Modular Optimisation

2026-04-30

AI Summary

This work addresses the challenge of concurrently optimizing discrete topology and continuous parameters in mechanical linkage design—a task poorly handled by conventional methods. The authors propose a modular framework that integrates large language models with numerical optimizers, employing a symbolic lifting operator to translate simulation trajectories into interpretable qualitative descriptions. This enables off-the-shelf, unfinetuned language models (e.g., Llama 3.3 70B and Qwen3 series) to perform explainable reasoning and constraint diagnosis for linkage structures. The approach achieves up to a 68% reduction in geometric error and a 134% improvement in structural validity. Moreover, 78.6% of optimization iterations yield measurable enhancements, and fault diagnosis accuracy exceeds 56%, substantially outperforming traditional strategies.

Abstract

Designing mechanical linkages involves combinatorial topology selection and continuous parameter fitting. We show that language models can systematically improve linkage designs through symbolic representations. Language model agents explore discrete topologies while numerical optimisers fit continuous parameters. A symbolic lifting operator translates simulator trajectories into qualitative descriptors, motion labels, temporal predicates, and structural diagnostics that models interpret across iterative design cycles. Across six engineering-relevant motion targets and three open-source models (Llama 3.3 70B, Qwen3 4B, Qwen3 MoE 30B-A3B), the modular architecture reduces geometric error by up to 68% and improves structural validity by up to 134% over monolithic baselines. Critically, 78.6% of iterative refinement trajectories show measurable improvement, with the system correctly diagnosing overconstraint (56.3%) and underconstraint (35.6%) failure modes and proposing grounded corrections. Models across all three families acquire interpretable mechanical reasoning strategies without fine-tuning, demonstrating that principled symbolic abstraction bridges generative AI and the numerical precision required for engineering design.

Problem

• mechanical linkage design
• combinatorial topology
• continuous parameter optimization
• symbolic representation
• engineering design

Innovation

• symbolic reflection
• modular optimisation
• mechanical linkage design
• language model agents
• symbolic lifting