# POET: Power-Oriented Evolutionary Tuning for LLM-Based RTL PPA Optimization

arXiv: https://arxiv.org/abs/2603.19333 · HTML: https://mlsanity.com/paper/2603.19333/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 33 references that are in the map.
- First posted on arXiv 19 Mar 2026.

## Builds on (most cited first, 11 of 11)

- [VerilogEval: Evaluating Large Language Models for Verilog Code Generation](https://mlsanity.com/paper/2309.07544/) (2023)
- [Revisiting VerilogEval: A Year of Improvements in Large-Language Models for Hardware Code Generation](https://mlsanity.com/paper/2408.11053/) (2024)
- [A Survey of Research in Large Language Models for Electronic Design Automation](https://mlsanity.com/paper/2501.09655/) (2025)
- [TS-Reasoner: Domain-Oriented Time Series Inference Agents for Reasoning and Automated Analysis](https://mlsanity.com/paper/2410.04047/) (2024)
- [VFlow: Discovering Optimal Agentic Workflows for Verilog Generation](https://mlsanity.com/paper/2504.03723/) (2025)
- [SymRTLO: Enhancing RTL Code Optimization with LLMs and Neuron-Inspired Symbolic Reasoning](https://mlsanity.com/paper/2504.10369/) (2025)
- [VeriMoA: A Mixture-of-Agents Framework for Spec-to-HDL Generation](https://mlsanity.com/paper/2510.27617/) (2025)
- [Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2601.21919/) (2026)
- [The Dawn of Agentic EDA: A Survey of Autonomous Digital Chip Design](https://mlsanity.com/paper/2512.23189/) (2025)
- [EvolVE: Evolutionary Search for LLM-based Verilog Generation and Optimization](https://mlsanity.com/paper/2601.18067/) (2026)
- [A Structure-Aware Framework for Learning Device Placements on Computation Graphs](https://mlsanity.com/paper/2405.14185/) (2024)

## Built on by (most cited first, 11 of 11)

- [Dr. RTL: Autonomous Agentic RTL Optimization through Tool-Grounded Self-Improvement](https://mlsanity.com/paper/2604.14989/) (2026)
- [RaMem: Contextual Reinstatement for Long-term Agentic Memory](https://mlsanity.com/paper/2606.22844/) (2026)
- [COEVO: Co-Evolutionary Framework for Joint Functional Correctness and PPA Optimization in LLM-Based RTL Generation](https://mlsanity.com/paper/2604.15001/) (2026)
- [ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling](https://mlsanity.com/paper/2606.24437/) (2026)
- [ARES: Adaptive Reasoning-Effort Steering for PPA- and Cost-Aware RTL Optimization with LLM Agents](https://mlsanity.com/paper/2607.27879/) (2026)
- [SynCo: Data Synthesis Co-Training for Self-Evolving LLMs via Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2610.11345/) (2026)
- [TimeMM: Time-as-Operator Spectral Filtering for Dynamic Multimodal Recommendation](https://mlsanity.com/paper/2604.26247/) (2026)
- [AUTOGATE: Automated Clock Gating via Toggling-Aware LLM-based RTL Rewriting](https://mlsanity.com/paper/2606.17461/) (2026)
- [ChipMEM: Verification-Grounded Memory for EDA Agents](https://mlsanity.com/paper/2609.27067/) (2026)
- [BEHAVE: Functional Behavior Modeling Enables Self-Improving Agents for Hardware Design and Verification](https://mlsanity.com/paper/2609.34785/) (2026)
- [Beyond Final Accuracy: Auditing Communication in LLM Multi-Agent Systems](https://mlsanity.com/paper/2610.01042/) (2026)

## Abstract (arXiv)

Applying large language models (LLMs) to RTL code optimization for improved power, performance, and area (PPA) faces two key challenges: ensuring functional correctness of optimized designs despite LLM hallucination, and systematically prioritizing power reduction within the multi-objective PPA trade-off space. We propose POET (Power-Oriented Evolutionary Tuning), a framework that addresses both challenges. For functional correctness, POET introduces a differential-testing-based testbench generation pipeline that treats the original design as a functional oracle, using deterministic simulation to produce golden references and eliminating LLM hallucination from the verification process. For PPA optimization, POET employs an LLM-driven evolutionary mechanism with non-dominated sorting, power-first intra-level ranking, and proportional survivor selection to steer the search toward the low-power region of the Pareto front without manual weight tuning. Evaluated on the RTL-OPT benchmark across 40 diverse RTL designs, POET achieves 100% functional correctness, the best power on all 40 designs, and competitive area and delay improvements.