IterSIMP-σ: Evaluating LLM-Assisted Spatial Interventions in Stress-Aware Topology Optimization

2026-05-18

AI Summary

This work investigates whether multimodal large language models (MLLMs) can serve as spatial intervention modules to assist stress-aware topology optimization. To this end, the authors propose the IterSIMP-σ framework, which wraps a closed-loop system around a conventional SIMP optimizer, integrating a deterministic stress channel, a gated evaluator, and a hybrid MLLM/rule-based interpreter. At each iteration, the framework generates verifiable soft-density seed interventions based on density and stress fields. This study represents the first effort to incorporate MLLMs into topology optimization as a spatial proposal mechanism while ensuring full verifiability throughout the process. Experiments on 2D benchmarks show that the MLLM-based approach achieves a marginally lower geometric mean compliance (−1.9%, not statistically significant); under fixed volume constraints, it attains fully feasible solutions in 25 out of 44 runs, with an average seed-to-hotspot distance of 0.221, demonstrating feasibility albeit without significant advantage over deterministic strategies.

Abstract

This paper studies whether multimodal large language models (LLMs) can serve as inspectable spatial proposal modules for stress-aware topology optimization. IterSIMP-σ keeps the SIMP optimizer as a compliance-minimizing finite-element solver and places a deterministic stress pass, gate evaluator, and hybrid LLM/rule interpreter around it. After each solve, density and von Mises stress fields are rendered; the interpreter proposes ranked spatial interventions; and deterministic safeguards accept, reject, or stop each action. The main action is a soft density seed, where selected elements are initialized at elevated density before the next solve but remain free under the optimality-criteria update. We evaluate the loop on a 16-problem 2D controller-policy benchmark, a six-problem exploratory 3D extension, passive-solid and input ablations, stress-threshold sensitivity, and a fixed-volume attribution study comparing LLM proposals with deterministic max-stress hotspot seeding, random stress-region seeding, and rule-based control. The 2D controller-policy benchmark shows a small retained-compliance difference (1.9% lower geometric mean for the soft-seed LLM), but this diagnostic is not statistically significant (W = 33, two-sided p = 0.382) and is not a fixed-volume feasible-final comparison. In the fixed-volume study, the LLM condition completed 44/48 attempted evaluations; 25/44 completed evaluations produced all-gate-passing retained states. Feasible-final scoring against rule-based control is split 4/4/1, and deterministic exact-hotspot seeding remains competitive. Accepted LLM spatial actions with per-step records have mean normalized seed-to-hotspot distance 0.221. The results support IterSIMP-σ as an inspectable LLM-assisted design-automation framework for spatial interventions, not yet as evidence that LLM visual reasoning improves stress-constrained optimization.

Problem

• topology optimization
• stress-aware design
• large language models
• spatial interventions
• multimodal LLMs

Innovation

• LLM-assisted design
• spatial intervention
• stress-aware topology optimization
• inspectable automation
• soft density seeding