LLMs and Knowledge Graphs Enhance SysML v2 Semantic Fault Localization

Researchers have developed a novel framework to automatically detect and fix semantic errors in SysML v2 models, which are not caught by traditional compilers. This system integrates a fine-tuned Small Language Model (SLM) with a domain-specific knowledge graph. The knowledge graph encodes rules for physical compatibility between system elements and aids in generating synthetic data for training the SLM. The framework was tested in the vehicle systems domain, demonstrating a significant improvement in semantic fault repair rates, from under 3% to over 91%, by outputting unified diff patches for engineer review. AI

IMPACT This framework could significantly improve the efficiency and accuracy of system design verification by automating the detection and repair of complex semantic errors.

RANK_REASON The cluster contains an academic paper detailing a new framework and methodology for semantic fault localization in SysML v2 models. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• DeepSeek-Coder-6.7B
• knowledge graph
• Qwen2.5-Coder-1.5B
• Small Language Model
• SysML v2

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →