Search papers, labs, and topics across Lattice.
This paper introduces StructureClaw, a novel artifact-centered workbench for LLM agents that enhances the evaluation of structural engineering workflows by ensuring that all interdependent artifacts are correctly generated and validated. By implementing an executable benchmark, StructureClaw-Bench, which includes 150 controlled scenarios, the authors demonstrate a significant improvement in success rates from 56.8% to 88.6% when using a full automatic workflow compared to a generic-skill baseline. The study highlights critical challenges in handling invalid inputs and maintaining consistency in structural model reconstruction, emphasizing the importance of comprehensive evaluation in engineering applications.
Artifact-centered evaluations reveal that LLM agents can achieve an 88.6% success rate in structural engineering tasks, but still struggle with invalid inputs and model consistency.
Addressing a structural-engineering request requires more than a single answer; it requires a chain of interdependent artifacts: interpreted requirements, a computable model, validation records, solver outputs, code-check records, and a final report. Evaluations centered on question answering or script generation rarely verify this complete evidence chain and may therefore reward fluent outputs even when the underlying engineering workflow is incomplete, internally inconsistent, or non-executable. To address this limitation, we present StructureClaw, an artifact-centered workbench in which LLM agents operate through governed engineering skills, typed tools, shared artifact state, and local analysis backends. We also introduce StructureClaw-Bench, an executable benchmark of 150 controlled scenarios spanning standard workflow execution, interactive robustness, and multimodal structural-model reconstruction. A scenario succeeds only when all required artifact- and execution-level assertions pass in a single run. Across ten agent-model configurations, each evaluated on the same 50 standard cases, the average Success Rate rises from 56.8% with the generic-skill baseline to 88.6% with the full automatic workflow. The interactive and multimodal evaluations identify two prominent remaining challenges: safe handling of invalid numerical inputs and fixture-consistent reconstruction of structural models. These findings show that artifact-centered evaluation can expose workflow-level failures that are difficult to identify from final responses alone, providing a more rigorous basis for evaluating and improving structural-engineering agents. The code and benchmark are available at https://github.com/structureclaw/structureclaw.
