Metraipolis › archive › dc_c854407ad013

Grounded and Faithful P&ID Reasoning: Constraining Vision-Language Models with Recovered Evidence Graphs — arXiv:2609.05880

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.05880
License (as archived)
CC0-1.0
Archived
2026-10-16T13:56:24.615Z
SHA-256 of the archived text
bbe1bf2a350e23b391d567a6ddee924c9633fd50d0f4fb8da53905436a495ce9
Stable URL of this page
https://metraipolis.com/documento/dc_c854407ad013

How to cite

Grounded and Faithful P&ID Reasoning: Constraining Vision-Language Models with Recovered Evidence Graphs — arXiv:2609.05880 — arXiv (en), https://arxiv.org/abs/2609.05880 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c854407ad013 (Metraipolis archive dc_c854407ad013, sha256 bbe1bf2a350e23b391d567a6ddee924c9633fd50d0f4fb8da53905436a495ce9)
BibTeX
@misc{dcc854407ad013,
  title = {Grounded and Faithful P&ID Reasoning: Constraining Vision-Language Models with Recovered Evidence Graphs — arXiv:2609.05880},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.05880},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c854407ad013; sha256 bbe1bf2a350e23b391d567a6ddee924c9633fd50d0f4fb8da53905436a495ce9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05880. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Piping and Instrumentation Diagrams (P&IDs) are the authoritative maps of process plants: isolation, maintenance, and HAZOP decisions depend on what connects to what. Vision-language models describe these sheets fluently, yet they often invent or miss process connections---and an invented or missed link can reverse an isolation or reachability call, so a plant decision cannot trust a fluent answer that was never checked against the linework. We instead recover an explicit graph of the drawing---its symbols, the process connections between them, and the tags that name them---and then require the model to answer only by querying that graph through seven read-only operators, so a topology claim is returned only when it cites the query results that support it. On TopoPID-VQA, a new suite of 3000 topology questions over these sheets, Graph-Grounded Harness (Ours) raises exact match accuracy from 36.7--41.3% under image-only prompting to 74.3--76.0% for Qwen3-VL-4B, Qwen3-VL-8B, and Gemma-4-E4B. It does so on an imperfect substrate: on Digitize-PID dataset the recovered graph scores F1 0.742 on exact process connections, and 0.801 once symbols and tags are pooled in. The residual errors track that gap---grounding pays off where the recovered graph is right, and perception error still breaks topology questions where it is not.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.05880 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T13:39:42.819Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).