We present a framework in which a large language model (LLM) acts as an online adaptive controller for SIMP topology optimization, replacing conventional fixed-schedule continuation with real-time, state-conditioned parameter decisions. At every $k$-th iteration, the LLM receives a structured observation$-$current compliance, grayness index, stagnation counter, checkerboard measure, volume fraction, and budget consumption$-$and outputs numerical values for the penalization exponent $p$, projection sharpness $β$, filter radius $r_{\min}$, and move limit $δ$ via a Direct Numeric Control interface. A hard grayness gate prevents premature binarization, and a meta-optimization loop uses a second LLM pass to tune the agent's call frequency and gate threshold across runs. We benchmark the agent against four baselines$-$fixed (no-continuation), standard three-field continuation, an expert heuristic, and a schedule-only ablation$-$on three 2-D problems (cantilever, MBB beam, L-bracket) at $120\!\times\!60$ resolution and two 3-D problems (cantilever, MBB beam) at $40\!\times\!20\!\times\!10$ resolution, all run for 300 iterations. A standardized 40-iteration sharpening tail is applied from the best valid snapshot so that compliance differences reflect only the exploration phase. The LLM agent achieves the lowest final compliance on every benchmark: $-5.7\%$ to $-18.1\%$ relative to the fixed baseline, with all solutions fully binary. The schedule-only ablation underperforms the fixed baseline on two of three problems, confirming that the LLM's real-time intervention$-$not the schedule geometry$-$drives the gain. Code and reproduction scripts will be released upon publication.

翻译：我们提出一种框架，其中大型语言模型（LLM）作为SIMP拓扑优化的在线自适应控制器，取代传统的固定调度延续，实现基于实时状态条件的参数决策。在每第$k$次迭代时，LLM接收结构化观测数据——当前柔顺度、灰度指数、停滞计数器、棋格度量、体积分数和预算消耗——并通过直接数值控制接口输出惩罚因子$p$、投影锐度$\beta$、滤波半径$r_{\min}$及移动极限$\delta$的数值。硬灰度门控机制防止过早二值化，元优化循环通过第二次LLM调用调整代理的调用频率和跨运行的阈值门限。我们以三个二维问题（悬臂梁、MBB梁、L型支架，分辨率$120\!\times\!60$）和两个三维问题（悬臂梁、MBB梁，分辨率$40\!\times\!20\!\times\!10$）为基准，在300次迭代下将代理与四种基线方法（固定无延续、标准三场延续、专家启发式及仅调度消融）进行对比。通过从最佳有效快照施加标准化40次迭代锐化尾段，确保柔顺度差异仅反映探索阶段。LLM代理在所有基准测试中均达到最低最终柔顺度：相对固定基线降低$-5.7\%$至$-18.1\%$且所有解完全二值化。仅调度消融在两个问题中表现差于固定基线，证实LLM的实时干预——而非调度几何结构——是性能提升的关键。代码与复现脚本将于发表时公开。