This page contains links to resources which may aid in creating diagrams or researching within the NN diagrams domain.
There are three papers at Diagrams 2021 relating directly to NN system diagrams [4,6,7].
There has been a discussion of different options for Rethinking Machine Learning publications as diagrams .
Interviews conducted with NN users  are available as full transcripts.
Supporting tables for  can be found here.
Examples of diagrams taken from EMNLP 2017, ACL 2018 and COLING 2018. These are included to show the heterogeneity of the diagrams used in recent AI literature.
Net2Vis research , done as part of creating software to automatically generate diagrams from Keras code, also resulted in a diagram corpus being created, available in the supplementary materials. There are further corpus-based insights and additional data included.
A corpus of all ACL2017 system diagrams, manually extracted as part of  is here.
It is common to use pdffigures2.0 for extracting corpora. Beware the known limitations which are documented!
For those interested in pedagogical and fine-grained neural architectural representation, nice and well-cited schematic diagrams of Neural Network Architectures can be found here: http://www.asimovinstitute.org/neural-network-zoo/
For schematics of NN architectures (distinct from NN system architectures) , there is a useful tool at http://alexlenail.me/NN-SVG/ 
A collection of NN diagramming tools can be found at https://github.com/ashishpatel26/Tools-to-Design-or-Visualize-Architecture-of-Neural-Network
A first version of the DIAL Specification can be found on arXiv.
In these initial stages, DIAL is built using IPE, chosen for its LaTeX support.
It is recommended to insert DIAL images into LaTeX as .eps files using \includegraphics.
 Marshall, Guy, and André Freitas. “The Diagrammatic AI Language (DIAL): Version 0.1.” arXiv preprint arXiv:1812.11142 (2018).
 LeNail, Alexander. “NN-SVG: Publication-Ready Neural Network Architecture Schematics.” Journal of Open Source Software 4 (2019): 33
 A. Bäuerle, C. van Onzenoodt and T. Ropinski, “Net2Vis – A Visual Grammar for Automatically Generating Publication-Tailored CNN Architecture Visualizations,” in IEEE Transactions on Visualization and Computer Graphics, vol. 27, no. 6, pp. 2980-2991, 1 June 2021, doi: 10.1109/TVCG.2021.3057483.
 Marshall, Guy Clarke, Caroline Jay, and André Freitas. “Number and quality of diagrams in scholarly publications is associated with number of citations.” International Conference on Theory and Application of Diagrams. Springer, Cham, 2021.
 Marshall, Guy Clarke, André Freitas, and Caroline Jay. “How researchers use diagrams in communicating neural network systems.” arXiv preprint arXiv:2008.12566 (2020).
 Marshall, Guy Clarke, Caroline Jay, and André Freitas. “Understanding scholarly neural network system diagrams through application of VisDNA.” International Conference on Theory and Application of Diagrams. Springer, Cham, 2021.
 Marshall, Guy Clarke, Caroline Jay, and André Freitas. “Structuralist analysis for neural network system diagrams.” International Conference on Theory and Application of Diagrams. Springer, Cham, 2021.
 Marshall, Guy Clarke, Caroline Jay, and André Freitas. “Diagrammatic summaries for neural architectures.” Beyond static papers: Rethinking how we share scientific understanding in ML-ICLR 2021 workshop. 2021.