Deja, Marek · Nov 5, 2026
Marek Deja presents tok, a plain-text notation for BPMN and DMN
Markdown · JSON · Original source
Marek Deja presents tok as a plain-text notation for BPMN process models and DMN decisions. An engine checks tok against both standards, compiles it to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol. In a comparison of 893 files exported by 37 tools, the paper reports that tok models use fewer tokens than BPMN XML. Its checker signaled more injected faults than XML Schema with bpmnlint. Four language models, given a one-page reference, wrote valid tok for 85.7 to 100% of process descriptions from two datasets; with a skill and one checker tool, the reported range was 97.6 to 100%. The paper also reports that the strongest model used fewer tokens writing tok than BPMN XML. The results present tok as a format language models can write, with an engine checking the documents. They come from one reported study; performance beyond the tested files, descriptions, models and tools remains unknown.
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
cs.AI · Article
Source details
Original source date: Nov 5, 2026 (2026-11-05).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-08T20:17:34.705Z. First observed: 2026-11-06T05:10:33.934Z. Retrieved: 2026-11-08T20:10:53.560Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The evidence provided is the abstract, not the full paper; details of the study methods and results are not available here. The abstract reports one study, with 87 process descriptions from two datasets and four language models; it does not establish performance beyond the tested materials and setup.
Open question
How would tok’s reported results change when tested on process descriptions, models, and tools beyond those in this study?
Source-stated
Marek Deja describes tok as a plain-text notation for BPMN 2.0.2 process models and DMN 1.5 decisions.
tok is a plain-text notation for Business Process Model and Notation (BPMN) 2.0.2 process models and Decision Model and Notation (DMN) 1.5 decisions.
Source-stated
The tok engine checks documents against both standards, compiles them to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol.
One engine checks a tok document against both standards, compiles the document to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol.
Source-stated
In the reported comparison, tok models used 2.2 times fewer tokens than the smallest BPMN XML files and 7.7 times fewer than files as exported.
A tok model takes 2.2 times fewer tokens than the smallest BPMN XML of 893 files exported by 37 tools and 7.7 times fewer than the files as exported.
Source-stated
The checker signaled 98.2% of injected faults, compared with 85.8% for XML Schema with bpmnlint.
The checker signals 98.2% of 1,411 injected faults, against 85.8% for the XML Schema with bpmnlint.
Source-stated
Four language models wrote valid tok for 85.7 to 100% of process descriptions from two datasets using a one-page reference; with a lean skill and one checker tool, the reported range was 97.6 to 100%.
Four language models new to tok write valid tok from a one-page reference for 85.7 to 100% of 87 process descriptions from two datasets
Source-stated
The paper reports one study, including tests using 87 process descriptions from two datasets; results beyond the tested materials and setup remain unknown.
The paper classifies the notation as an alternative concrete syntax of the BPMN and DMN metamodels and reports one study.
Source and contributor context
Deja, Marek · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
No suitable media was found by the supported page extraction.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Lean
Lean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Editorial provenance
Version 1. Publication ID: 85b50e98-b87a-4d5e-84be-fdd45d276b2b.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-08T20:17:34.705Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: b08e634b5a90dcb931df83013d67a264050213e8ff583dcb6f09bd01253d90dc. Source text hash: a72169701b1cfcfa5967c151e249ed5899515589cacc1ccee549be4de2851a15.
Revision and source-check history
Version 1 · 2026-11-08T20:17:34.705Z
Marek Deja presents tok, a plain-text notation for BPMN and DMN
Marek Deja presents tok as a plain-text notation for BPMN process models and DMN decisions. An engine checks tok against both standards, compiles it to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol. In a comparison of 893 files exported by 37 tools, the paper reports that tok models use fewer tokens than BPMN XML. Its checker signaled more injected faults than XML Schema with bpmnlint. Four language models, given a one-page reference, wrote valid tok for 85.7 to 100% of process descriptions from two datasets; with a skill and one checker tool, the reported range was 97.6 to 100%. The paper also reports that the strongest model used fewer tokens writing tok than BPMN XML. The results present tok as a format language models can write, with an engine checking the documents. They come from one reported study; performance beyond the tested files, descriptions, models and tools remains unknown.
Original source
cs.AI · Article
Source details
Original source date: Nov 5, 2026 (2026-11-05).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-08T20:17:34.705Z. First observed: 2026-11-06T05:10:33.934Z. Retrieved: 2026-11-08T20:10:53.560Z.
Limits
The evidence provided is the abstract, not the full paper; details of the study methods and results are not available here. The abstract reports one study, with 87 process descriptions from two datasets and four language models; it does not establish performance beyond the tested materials and setup.
Open question
How would tok’s reported results change when tested on process descriptions, models, and tools beyond those in this study?
Source-stated
Marek Deja describes tok as a plain-text notation for BPMN 2.0.2 process models and DMN 1.5 decisions.
tok is a plain-text notation for Business Process Model and Notation (BPMN) 2.0.2 process models and Decision Model and Notation (DMN) 1.5 decisions.
Source-stated
The tok engine checks documents against both standards, compiles them to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol.
One engine checks a tok document against both standards, compiles the document to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol.
Source-stated
In the reported comparison, tok models used 2.2 times fewer tokens than the smallest BPMN XML files and 7.7 times fewer than files as exported.
A tok model takes 2.2 times fewer tokens than the smallest BPMN XML of 893 files exported by 37 tools and 7.7 times fewer than the files as exported.
Source-stated
The checker signaled 98.2% of injected faults, compared with 85.8% for XML Schema with bpmnlint.
The checker signals 98.2% of 1,411 injected faults, against 85.8% for the XML Schema with bpmnlint.
Source-stated
Four language models wrote valid tok for 85.7 to 100% of process descriptions from two datasets using a one-page reference; with a lean skill and one checker tool, the reported range was 97.6 to 100%.
Four language models new to tok write valid tok from a one-page reference for 85.7 to 100% of 87 process descriptions from two datasets
Source-stated
The paper reports one study, including tests using 87 process descriptions from two datasets; results beyond the tested materials and setup remain unknown.
The paper classifies the notation as an alternative concrete syntax of the BPMN and DMN metamodels and reports one study.
Source and contributor context
Deja, Marek · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
No suitable media was found by the supported page extraction.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Lean
Lean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Editorial provenance
Version 1. Publication ID: 85b50e98-b87a-4d5e-84be-fdd45d276b2b.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-08T20:17:34.705Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: b08e634b5a90dcb931df83013d67a264050213e8ff583dcb6f09bd01253d90dc. Source text hash: a72169701b1cfcfa5967c151e249ed5899515589cacc1ccee549be4de2851a15.
uncertain · 2026-11-08T20:17:34.705Z · anthropic/claude-opus-5.5
Publication: 85b50e98-b87a-4d5e-84be-fdd45d276b2b
attribution — pass: The draft correctly attributes tok and the reported results to the paper's sole listed author, Marek Deja. Results are framed as what the paper reports, not as established facts.
Authors:Marek Deja
summary — uncertain: Two gaps. First, central terms are never explained for a lay reader. The draft does not expand or describe BPMN and DMN (standards for modeling business processes and decisions), tokens, XML, Model Context Protocol, bpmnlint or lean skill, so the significance is hard to grasp. Second, it omits a key condition: the same models wrote valid BPMN XML for every description, while valid tok ranged from 85.7 to 100%. Smallest repair: add that comparison and brief plain-language glosses or tooltips for BPMN, DMN and tokens.
write valid tok from a one-page reference for 85.7 to 100% of 87 process descriptions from two datasets and valid BPMN XML for every description
date — pass: The originalPublished date of 2026-11-05 matches the arXiv v1 submission date. It is not the later feed date.
[Submitted on 5 Nov 2026]
support — pass: Every stated figure and claim matches the abstract: the 2.2x and 7.7x token reductions, the 98.2% vs 85.8% fault detection, the 85.7 to 100% and 97.6 to 100% validity ranges, and the strongest model using fewer tokens. The draft's limitation about untested settings is stated as unknown rather than asserted.
The checker signals 98.2% of 1,411 injected faults, against 85.8% for the XML Schema with bpmnlint.
media — pass: The selected media is the verified direct link to the arXiv abstract page, which is the original source.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Lean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Explore LeanarXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXivLean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Explore LeanarXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXiv