Retrieve, Schedule, Reflect: LLM Agents for Chip QoR Optimization
This work addresses the need for automated, expert-level chip design optimization to reduce manual intervention and improve efficiency in the semiconductor industry, representing a novel application rather than an incremental improvement.
The paper tackles the problem of automating chip design optimization by proposing an LLM agent framework that schedules optimizations through interaction with EDA tools, achieving 10% greater timing improvement, reduced power and area, and a 4x speedup compared to black-box methods.
Modern chip design requires multi-objective optimization of timing, power, and area under stringent time-to-market constraints. Although powerful optimization algorithms are integrated into EDA tools, achieving high QoR hinges on effective long-horizon scheduling, which relies heavily on manual expert intervention. To address this issue and automate chip design, we propose an agentic LLM framework that schedules chip optimizations through direct interaction with EDA tools. The agent is grounded in natural language expertise expressed as a search tree through retrieval-augmented generation (RAG). We further improve scheduling quality with Pareto-driven QoR feedback through language reflection. Experimental results show that, compared with black-box search methods such as reinforcement learning, our framework achieves 10% greater timing improvement while consuming less power and area, with more than 4x speedup. The post-optimization QoR is also comparable to that achieved by human experts. Finally, the agent supports customized tasks expressed in natural language, enabling preferential QoR trade-offs. The code and chip design data will be publicly available at https://github.com/YiKangOY/Open-LLM-ECO.