Search papers, labs, and topics across Lattice.
CODMAS, a novel framework for automated RTL optimization, leverages a dialectic multi-agent system composed of an Articulator, Hypothesis Partner, Domain-Specific Coding Agent (DCA), and Code Evaluation Agent (CEA) to iteratively refine Verilog code. The Articulator and Hypothesis Partner engage in structured dialectic reasoning to guide the DCA in generating architecture-aware Verilog edits, which are then verified by the CEA for syntax, functionality, and PPA metrics. Evaluated on the newly introduced RTLOPT benchmark, CODMAS achieves significant improvements in critical path delay (~25% reduction for pipelining) and power consumption (~22% reduction for clock gating) compared to baseline methods.
A multi-agent system that mimics rubber-duck debugging slashes critical path delay by 25% and power consumption by 22% in RTL code, outperforming LLM-based baselines.
Optimizing Register Transfer Level (RTL) code is a critical step in Electronic Design Automation (EDA) for improving power, performance, and area (PPA). We present CODMAS (Collaborative Optimization via a Dialectic Multi-Agent System), a framework that combines structured dialectic reasoning with domain-aware code generation and deterministic evaluation to automate RTL optimization. At the core of CODMAS are two dialectic agents: the Articulator, inspired by rubber-duck debugging, which articulates stepwise transformation plans and exposes latent assumptions; and the Hypothesis Partner, which predicts outcomes and reconciles deviations between expected and actual behavior to guide targeted refinements. These agents direct a Domain-Specific Coding Agent (DCA) to generate architecture-aware Verilog edits and a Code Evaluation Agent (CEA) to verify syntax, functionality, and PPA metrics. We introduce RTLOPT, a benchmark of 120 Verilog triples (unoptimized, optimized, testbench) for pipelining and clock-gating transformations. Across proprietary and open LLMs, CODMAS achieves ~25% reduction in critical path delay for pipelining and ~22% power reduction for clock gating, while reducing functional and compilation failures compared to strong prompting and agentic baselines. These results demonstrate that structured multi-agent reasoning can significantly enhance automated RTL optimization and scale to more complex designs and broader optimization tasks.