SCIENCE UNDER AUTOMATION
Physics-in-the-Loop
Agents generate and revise CAD using explicit engineering tools and a supplied load case; the evaluation reports increased structural complexity and compile rate.
What this result supports
Computational verification is within the supplied model. Fabrication, commissioning and physical certification are outside the demonstrated workflow.
Evidence record: Peer-reviewed IJCAI engineering benchmark. Checked: 2026-10-21.
Where this evidence is used
Engineering
- Translate requirements into a coherent design: CAD benchmarks show progress with tool feedback and substantial failures on complete typed requirements.
- Generate and revise executable models: Agent loops generate designs and revise them using computational feedback.
- Fabricate and integrate the complete system: The cited design evaluations end with computational artifacts; real integration remains specialist work.
- Measure performance under real operating conditions: Simulation validation does not perform physical commissioning or field testing.