COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration

📰 ArXiv cs.AI

COSMO-Agent is a tool-augmented reinforcement learning framework that teaches LLMs to complete the closed-loop CAD-CAE process

advanced Published 8 Apr 2026
Action Steps
  1. Identify the CAD-CAE semantic gap in industrial design-simulation optimization
  2. Develop a tool-augmented reinforcement learning framework to teach LLMs
  3. Implement COSMO-Agent to complete the closed-loop CAD-CAE process
  4. Evaluate the performance of COSMO-Agent in various design optimization scenarios
Who Needs to Know This

This benefits cross-functional teams of engineers, designers, and researchers working on complex design optimization problems, as it enables them to automate the CAD-CAE process and improve design efficiency

Key Insight

💡 COSMO-Agent can automate the CAD-CAE process by teaching LLMs to translate simulation feedback into valid geometric edits

Share This
🤖 COSMO-Agent: a tool-augmented RL framework for closed-loop CAD-CAE optimization

Key Takeaways

COSMO-Agent is a tool-augmented reinforcement learning framework that teaches LLMs to complete the closed-loop CAD-CAE process

Full Article

Title: COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration

Abstract:
arXiv:2604.05547v1 Announce Type: new Abstract: Iterative industrial design-simulation optimization is bottlenecked by the CAD-CAE semantic gap: translating simulation feedback into valid geometric edits under diverse, coupled constraints. To fill this gap, we propose COSMO-Agent (Closed-loop Optimization, Simulation, and Modeling Orchestration), a tool-augmented reinforcement learning (RL) framework that teaches LLMs to complete the closed-loop CAD-CAE process. Specifically, we cast CAD generat
Read full paper → ← Back to Reads