Companion artefact to the MODELS 2026 (NIER) paper. Does epistemic metadata change how AI consumers answer derivation questions over an MBSE model? 3 LLMs x 120 judged answers over the public Apollo 11 SysML v2 model. doi:10.1145/3822455.3838783
provenance reproducibility traceability ai-safety systems-engineering apollo-11 mbse hallucination model-based-systems-engineering grounding digital-engineering llm-evaluation confabulation sysml-v2 research-artifact epistemic-adequacy ai-native-mbse governed-interfaces
-
Updated
Aug 18, 2026 - Python