Jul 2026
ARES: Adaptive Reasoning-Effort Steering for PPA- and Cost-Aware RTL Optimization with LLM Agents
A normalized dollar cost per LLM call reported alongside the figure of merit (FoM) is introduced, enabling fair comparison across effort levels and optimizers and finding the construction of the long-term memory matters little.
Stef Cuyckens, M. Jivanescu, Jun Yin et al.
· arXiv.org · 0 citations