Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• General-purpose LLMs fail to satisfy sequential, multi-gate constraints when generating executable simulation artifacts for thermal protection system design.limitation 85%
• AeroTherm-GPT is the first TPS-specialized LLM Agent for thermal protection system engineering workflows.method 90%
• The Constraint-Closed-Loop Generation (CCLG) framework organizes TPS artifact generation as an iterative workflow with five stages.method 90%
• The Constraint Dependency Graph (CDG) encodes empirical co-resolution structure among constraint categories to direct repair toward upstream fault candidates.technical 85%
• The upstream-priority mechanism achieves a Root-Cause Fix Efficiency of 4.16 versus 1.76 for flat-checklist repair.result 90%
• AeroTherm-GPT achieves 88.7% End-to-End Success Rate on HyTPS-Bench, a gain of +12.5 percentage points over the non-CDG ablation baseline.result 90%
• AeroTherm-GPT does not suffer from catastrophic forgetting on scientific reasoning and code generation tasks.result 85%
• Integrating LLMs into hypersonic thermal protection system design is bottlenecked by cascading constraint violations when generating executable simulation artifacts.technical 80%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card