{"slug": "statetune-transforming-llm-assisted-eda-flow-tuning-into-a-stateful-closed-loop", "title": "StateTune: Transforming LLM-Assisted EDA Flow Tuning into a Stateful, Closed-Loop Process", "summary": "Researchers introduced StateTune, a closed-loop LLM-assisted EDA flow tuning system with persistent optimization memory, which achieved the strongest final hypervolume on all six benchmark blocks in a Cadence industrial flow, outperforming five baselines including LLM+retrieval-augmented generation and preference-based Bayesian optimization. Ablation showed that removing persistent memory costs 58.5% of the hypervolume, and three-seed reproducibility showed coefficient of variation below 7% on five of six blocks.", "body_md": "arXiv:2608.23601v1 Announce Type: cross\nAbstract: EDA flow parameter tuning is critical for quality-of-results~(QoR), yet the parameter space is large, tightly coupled, and full evaluations are prohibitively expensive. Prior LLM-assisted tuners mainly use the LLM as an external proposer with transient working context; we instead present \\textbf{StateTune}, which reformulates LLM-assisted EDA tuning as a closed-loop, state-carrying process. Its optimizer state is a typed, evidence-gated \\emph{persistent optimization memory} that is updated by every evaluation and shared between candidate generation and budget allocation. On top of this optimizer state, an expected hypervolume improvement (EHVI)-guided, runtime-aware promotion policy ranks quick-stage candidates by expected Pareto frontier gain per unit of runtime cost. Evaluated on a Cadence industrial flow across six benchmark blocks (two technology nodes \\(\\times\\) three designs), against five baselines including LLM+retrieval-augmented generation (RAG) and preference-based Bayesian optimization (BO) tuners, StateTune achieves the strongest final hypervolume on all six benchmark blocks, showing a stable improvement in frontier quality across the full matrix; it also matches or surpasses the strongest baselines on worst negative slack (WNS), area, and power across the same set. Ablation shows persistent memory is the largest contributor: removing it costs 58.5\\% of the hypervolume. Dedicated analyses of evidence-gating sensitivity, memory poisoning, cross-design transfer, and three-seed reproducibility (CV\\,\\(<\\)\\,7\\% on five of six blocks) further validate the memory design.", "url": "https://wpnews.pro/news/statetune-transforming-llm-assisted-eda-flow-tuning-into-a-stateful-closed-loop", "canonical_source": "https://www.machinebrief.com/news/statetune-transforming-llm-assisted-eda-flow-tuning-into-a-s-hkz3", "published_at": "2026-08-26 04:00:00+00:00", "updated_at": "2026-08-26 06:13:54.652597+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research"], "entities": ["StateTune", "Cadence", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/statetune-transforming-llm-assisted-eda-flow-tuning-into-a-stateful-closed-loop", "markdown": "https://wpnews.pro/news/statetune-transforming-llm-assisted-eda-flow-tuning-into-a-stateful-closed-loop.md", "text": "https://wpnews.pro/news/statetune-transforming-llm-assisted-eda-flow-tuning-into-a-stateful-closed-loop.txt", "jsonld": "https://wpnews.pro/news/statetune-transforming-llm-assisted-eda-flow-tuning-into-a-stateful-closed-loop.jsonld"}}