Search papers, labs, and topics across Lattice.
This paper introduces SemaPLC, a verification-gated agent harness designed to enhance the reliability of PLC code generation by ensuring that generated logic integrates correctly into existing projects and operates as intended. By implementing a strict completion rule that requires external verification checks鈥攃overing specification, compilation, and runtime behavior鈥擲emaPLC achieves a mean verified pass rate of 72.6% across 117 independent POU tasks, outperforming all seven models tested. Notably, in a project-context evaluation of 65 tasks, SemaPLC demonstrated superior dynamic behavior scores, highlighting the critical importance of runtime execution in validating the functionality of generated control logic.
SemaPLC achieves a remarkable 52.2% dynamic behavior score, setting a new standard for verifying the operational integrity of PLC code generation.
Programmable logic controllers (PLCs) run industrial plants, and large language models can already generate independent program organization units (POUs) for them. Whether such logic integrates into an existing PLC project and then runs correctly has been checked only in limited tests. We present \textsc{SemaPLC}, a project-grounded and verification-gated agent harness assembled from conventional tools but governed by a strict completion rule. Rather than stopping when the model judges its own output adequate, \textsc{SemaPLC} declares a task complete only when logged external checks confirm it. Those checks cover the specification, the compilation, and the behavior on a live runtime. On 117 independent-POU tasks matching existing benchmarks, it attains the highest strict verified pass rate on all seven models (72.6\% mean). On a project-context track of 65 tasks whose generated logic must compile and run inside a real project, it attains the highest mean on integrated compilation, static behavior, and dynamic behavior. Of the three layers, dynamic behavior is the most revealing. We measure it by deploying the generated and the reference logic to a live PLC runtime and comparing their executed traces. All methods fall within 10 static points of one another, whereas dynamic scores separate them sharply, from 22.4 to 31.4 for the baselines against 52.2 for \textsc{SemaPLC}. Overall, our verification-gated harness raises the mean at every layer and most sharply at runtime. Execution, not static scoring, is the faithful test of whether generated control logic actually works. \textsc{SemaPLC} is open-sourced at https://github.com/midea-ai/SemaPLC.
