HYPERHEURIST: A Simulated Annealing-Based Control Framework for LLM-Driven Code Generation in Optimized Hardware Design
For hardware designers, this framework improves the reliability and efficiency of LLM-assisted RTL design by ensuring functional validity before PPA optimization.
HYPERHEURIST uses simulated annealing to iteratively refine LLM-generated RTL hardware designs, filtering for functional correctness before optimizing power-performance-area (PPA). Across eight benchmarks, it achieves more stable and repeatable optimization than single-shot LLM generation.
Large Language Models (LLMs) have shown promising progress for generating Register Transfer Level (RTL) hardware designs, largely because they can rapidly propose alternative architectural realizations. However, single-shot LLM generation struggles to consistently produce designs that are both functionally correct and power-efficient. This paper proposes HYPERHEURIST, a simulated annealing-based control framework that treats LLM-generated RTL as intermediate candidates rather than final designs. The suggested system not only focuses on functionality correctness but also on Power-Performance-Area (PPA) optimization. In the first phase, RTL candidates are filtered through compilation, structural checks, and simulation to identify functionally valid designs. PPA optimization is restricted to RTL designs that have already passed compilation and simulation. Evaluated across eight RTL benchmarks, this staged approach yields more stable and repeatable optimization behavior than single-pass LLM-generated RTL.