Automating the semantic validation of SysML v2models through a multi-agent framework. Jose Olivert-Iserte, E. C. & Álvarez-Rodríguez, J. M. 2026. ![Automating the semantic validation of SysML v2models through a multi-agent framework [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Ensuring the semantic alignment between system models and natural language requirements is a critical bottleneck for the adoption of Model-Based Systems Engineering (MBSE). While SysML v2 offers enhanced expressiveness compared to SysML v1, including: (i) first-class support for behavioral and structural integration, (ii) improved typing and reuse mechanisms, (iii) formal semantics enabling precise model interpretation, existing validation toolsprimarily target syntactic correctness leaving a gap in automated semantic assessment. The present work proposes a deterministic multi-agent framework designed to automate the semantic validation of SysML v2 models. Unlike traditional monolithic Large Language Model (LLM) approaches, the proposed framework decomposes the validation process into a specialized pipeline of five agents: model abstraction, requirement normalization, semantic judgement (grounded in the requirements quality characteristics defined in the ISO/IEC/IEEE 29148:2018 ”Systems and software engineering — Life cycle processes — Requirements engineering”), metric aggregation, and actionable feedback generation. The approach is evaluated through an empirical study using a dataset of 14 SysML v2 models and 303 requirements derived from official specifications. A comparative analysis is performed using two different LLM backends: a highefficiency proprietary model (GPT-4o-mini) and an open-weight reasoning model (Deepseek-R1 ). Results show that the multi-agent orchestrationreliably identifies semantic mismatches, incomplete realizations, and omissions across varying levels of system complexity, ranging from small componentlevel models (M1–M5) to large system-of-systems architectures involving multiple interacting subsystems. The findings show that while the openweight model exhibits a more permissive validation stance, the proprietary backend provides a conservative and highly precise assessment suitable forsafety-critical domains. This work contributes to the automation of auditable model quality assessment, providing a scalable path for iterative model refinement in modern software and systems engineering.
@misc{olivert_iserte_automating_2026,
title = {{Automating the semantic validation of {SysML} v2models through a multi-agent framework}},
url = {https://www.ssrn.com/abstract=6553831},
doi = {10.2139/ssrn.6553831},
abstract = {Ensuring the semantic alignment between system models and natural language requirements is a critical bottleneck for the adoption of Model-Based Systems Engineering (MBSE). While SysML v2 offers enhanced expressiveness compared to SysML v1, including: (i) first-class support for behavioral and structural integration, (ii) improved typing and reuse mechanisms, (iii) formal semantics enabling precise model interpretation, existing validation toolsprimarily target syntactic correctness leaving a gap in automated semantic assessment. The present work proposes a deterministic multi-agent framework designed to automate the semantic validation of SysML v2 models. Unlike traditional monolithic Large Language Model (LLM) approaches, the proposed framework decomposes the validation process into a specialized pipeline of five agents: model abstraction, requirement normalization, semantic judgement (grounded in the requirements quality characteristics defined in the ISO/IEC/IEEE 29148:2018 ”Systems and software engineering — Life cycle processes — Requirements engineering”), metric aggregation, and actionable feedback generation. The approach is evaluated through an empirical study using a dataset of 14 SysML v2 models and 303 requirements derived from official specifications. A comparative analysis is performed using two different LLM backends: a highefficiency proprietary model (GPT-4o-mini) and an open-weight reasoning model (Deepseek-R1 ). Results show that the multi-agent orchestrationreliably identifies semantic mismatches, incomplete realizations, and omissions across varying levels of system complexity, ranging from small componentlevel models (M1–M5) to large system-of-systems architectures involving multiple interacting subsystems. The findings show that while the openweight model exhibits a more permissive validation stance, the proprietary backend provides a conservative and highly precise assessment suitable forsafety-critical domains. This work contributes to the automation of auditable model quality assessment, providing a scalable path for iterative model refinement in modern software and systems engineering.},
urldate = {2026-05-13},
publisher = {SSRN},
author = {Jose Olivert-Iserte, Eduardo Cibri{\'{a}}n, Roy Mendieta and
Jos{\'{e}} Mar{\'{\i}}a {\'{A}}lvarez-Rodr{\'{\i}}guez },
year = {2026},
}
Downloads: 0
{"_id":"6dpZgqt3pTubkG2Pq","bibbaseid":"joseolivertiserte-lvarezrodrguez-automatingthesemanticvalidationofsysmlv2modelsthroughamultiagentframework-2026","author_short":["Jose Olivert-Iserte, E. C.","Álvarez-Rodríguez, J. M."],"bibdata":{"bibtype":"misc","type":"misc","title":"Automating the semantic validation of SysML v2models through a multi-agent framework","url":"https://www.ssrn.com/abstract=6553831","doi":"10.2139/ssrn.6553831","abstract":"Ensuring the semantic alignment between system models and natural language requirements is a critical bottleneck for the adoption of Model-Based Systems Engineering (MBSE). While SysML v2 offers enhanced expressiveness compared to SysML v1, including: (i) first-class support for behavioral and structural integration, (ii) improved typing and reuse mechanisms, (iii) formal semantics enabling precise model interpretation, existing validation toolsprimarily target syntactic correctness leaving a gap in automated semantic assessment. The present work proposes a deterministic multi-agent framework designed to automate the semantic validation of SysML v2 models. Unlike traditional monolithic Large Language Model (LLM) approaches, the proposed framework decomposes the validation process into a specialized pipeline of five agents: model abstraction, requirement normalization, semantic judgement (grounded in the requirements quality characteristics defined in the ISO/IEC/IEEE 29148:2018 ”Systems and software engineering — Life cycle processes — Requirements engineering”), metric aggregation, and actionable feedback generation. The approach is evaluated through an empirical study using a dataset of 14 SysML v2 models and 303 requirements derived from official specifications. A comparative analysis is performed using two different LLM backends: a highefficiency proprietary model (GPT-4o-mini) and an open-weight reasoning model (Deepseek-R1 ). Results show that the multi-agent orchestrationreliably identifies semantic mismatches, incomplete realizations, and omissions across varying levels of system complexity, ranging from small componentlevel models (M1–M5) to large system-of-systems architectures involving multiple interacting subsystems. The findings show that while the openweight model exhibits a more permissive validation stance, the proprietary backend provides a conservative and highly precise assessment suitable forsafety-critical domains. This work contributes to the automation of auditable model quality assessment, providing a scalable path for iterative model refinement in modern software and systems engineering.","urldate":"2026-05-13","publisher":"SSRN","author":[{"propositions":[],"lastnames":["Jose","Olivert-Iserte"],"firstnames":["Eduardo","Cibrián"],"suffixes":["Roy","Mendieta"]},{"firstnames":["José","María"],"propositions":[],"lastnames":["Álvarez-Rodríguez"],"suffixes":[]}],"year":"2026","bibtex":"@misc{olivert_iserte_automating_2026,\n\ttitle = {{Automating the semantic validation of {SysML} v2models through a multi-agent framework}},\n\turl = {https://www.ssrn.com/abstract=6553831},\n\tdoi = {10.2139/ssrn.6553831},\n\tabstract = {Ensuring the semantic alignment between system models and natural language requirements is a critical bottleneck for the adoption of Model-Based Systems Engineering (MBSE). While SysML v2 offers enhanced expressiveness compared to SysML v1, including: (i) first-class support for behavioral and structural integration, (ii) improved typing and reuse mechanisms, (iii) formal semantics enabling precise model interpretation, existing validation toolsprimarily target syntactic correctness leaving a gap in automated semantic assessment. The present work proposes a deterministic multi-agent framework designed to automate the semantic validation of SysML v2 models. Unlike traditional monolithic Large Language Model (LLM) approaches, the proposed framework decomposes the validation process into a specialized pipeline of five agents: model abstraction, requirement normalization, semantic judgement (grounded in the requirements quality characteristics defined in the ISO/IEC/IEEE 29148:2018 ”Systems and software engineering — Life cycle processes — Requirements engineering”), metric aggregation, and actionable feedback generation. The approach is evaluated through an empirical study using a dataset of 14 SysML v2 models and 303 requirements derived from official specifications. A comparative analysis is performed using two different LLM backends: a highefficiency proprietary model (GPT-4o-mini) and an open-weight reasoning model (Deepseek-R1 ). Results show that the multi-agent orchestrationreliably identifies semantic mismatches, incomplete realizations, and omissions across varying levels of system complexity, ranging from small componentlevel models (M1–M5) to large system-of-systems architectures involving multiple interacting subsystems. The findings show that while the openweight model exhibits a more permissive validation stance, the proprietary backend provides a conservative and highly precise assessment suitable forsafety-critical domains. This work contributes to the automation of auditable model quality assessment, providing a scalable path for iterative model refinement in modern software and systems engineering.},\n\turldate = {2026-05-13},\n\tpublisher = {SSRN},\n\tauthor = {Jose Olivert-Iserte, Eduardo Cibri{\\'{a}}n, Roy Mendieta and\n Jos{\\'{e}} Mar{\\'{\\i}}a {\\'{A}}lvarez-Rodr{\\'{\\i}}guez },\n\tyear = {2026},\n}\n\n\n","author_short":["Jose Olivert-Iserte, E. C.","Álvarez-Rodríguez, J. M."],"key":"olivert_iserte_automating_2026","id":"olivert_iserte_automating_2026","bibbaseid":"joseolivertiserte-lvarezrodrguez-automatingthesemanticvalidationofsysmlv2modelsthroughamultiagentframework-2026","role":"author","urls":{"Paper":"https://www.ssrn.com/abstract=6553831"},"metadata":{"authorlinks":{}}},"bibtype":"misc","biburl":"https://raw.githubusercontent.com/sa-mooc-uc3m/mooc-research-public/master/trunk/bibtex/chema.bib","dataSources":["KWj4zjrb2ijPcS3uM"],"keywords":[],"search_terms":["automating","semantic","validation","sysml","v2models","through","multi","agent","framework","jose olivert-iserte","álvarez-rodríguez"],"title":"Automating the semantic validation of SysML v2models through a multi-agent framework","year":2026}