Workshop co-allocated with EuroVis '26, Nottingham, England, 8 June 2026
Workshop description
Graph Drawing (GD) and Dimensionality Reduction (DR) share many methodological, theoretical, and practical challenges. Both aim to project complex, high-dimensional structures into interpretable low-dimensional representations, facing issues of scalability, ambiguity, and interpretability. Yet, despite these similarities, the two communities have evolved largely in parallel with limited exchange of ideas and evaluation practices.
This workshop seeks to bridge GD and DR by creating a space for discussion on how methods, concepts, and lessons from one domain can inspire advances in the other. Participants will explore common evaluation pitfalls, domain-expert usage scenarios, and reliability issues such as comparing real versus synthetic data. The workshop aims to build synergy between the two communities and produce outcomes such as a community statement on evaluation and a roadmap for future collaboration.
Title: Dimensionality Reduction, Machine Learning, and Graph Drawing
Abstract: The fields of graph drawing and dimensionality reduction have a long tradition. From the connection between Kamada and Kawai and Multidimensional Scaling there has been a close connection between these to fields. In this talk, I present a personalised retrospective on how these two fields have interacted and how we can interact in the age of data science and AI.
Biography: Daniel Archambault is a Professor of Visualisation and Data Science at Newcastle University. His research focuses on enabling visualisation and visual analytics systems to scale in the age of data science. He explores key problems in data science and AI, graph drawing, social and complex network analysis, and human-computer interaction, often within interdisciplinary contexts. His work spans the full data-to-human pipeline—from developing algorithms for visualising abstract data to conducting perceptual evaluations of these interfaces with users. His primary interests lie in visual analytics for machine learning and network visualisation.
Monday (8th June), half-day workshop (14:00–17:30):
14:00–14:05 — Welcoming and Introduction
14:05–15:30 — Paper Session: contributed short papers (8 minutes for each paper + 2 minutes for Q&A) (session chair: Claudio Linhares)
GNN-Based Inductive Dimensionality Reduction on UMAP Graphs, Florian Grötschla
Visualization and Evaluation of Multivariate Networks through Dimensionality Reduction and Graph Embeddings, Pedro A. M. Gagini
Readability vs. Faithfulness: Unveiling Correlations between Graph Aesthetics and DR Quality, Seokweon Jung
NodeSubstrates: Bridging Node-Link Diagrams with DR-Based Semantic Substrates, Sjoerd Vink
Bridging Graph Drawing and Dimensionality Reduction with Stochastic Stress Optimization, Jacob Miller
Multivariate Graph Layouts using NNP-NET, Ilan Hartskeerl
15:30–16:00 — Coffee Break
16:00–17:00 — Invited Keynote: Prof. Dr. Daniel Archambault - Title Dimensionality Reduction, Machine Learning, and Graph Drawing (followed by Q&A) (session chair: Andreas Kerren)
17:00–17:25 — Open Discussions: mapping synergies, challenges, and opportunities (session chair: Fernando Paulovich)
17:25–17:30 — Closing Remarks
The program combines presentations, interactive discussions, and collaborative activities to identify synergies, methodological parallels, and future research opportunities across GD and DR.
This collection of papers is available on: https://diglib.eg.org/handle/10.2312/evgdxdr20261000
Visualization and Evaluation of Multivariate Networks through Dimensionality Reduction and Graph Embeddings
Pedro A. M. Gagini¹, Rafael M. Martins², Amilcar Soares², Fernando V. Paulovich³, Claudio D. G. Linhares²
¹University of São Paulo, Brazil · ²Linnaeus University, Sweden · ³Eindhoven University of Technology, Netherlands
Readability vs. Faithfulness: Unveiling Correlations between Graph Aesthetics and DR Quality
Seokweon Jung¹ ², Min Hyeong Kim², Hyeon Jeon², Jinwook Seo²
¹KAIST, South Korea · ²Seoul National University, South Korea
NodeSubstrates: Bridging Node-Link Diagrams with DR-Based Semantic Substrates
Sjoerd Vink¹ ², Leonardo Christino², Michael Behrisch¹ ²
¹Utrecht University, Netherlands · ²GraphPolaris, Netherlands
GNN-Based Inductive Dimensionality Reduction on UMAP Graphs
Florian Grötschla¹, Simon van Wageningen², Alexandru Telea²
¹ETH Zurich, Switzerland · ²Utrecht University, Netherlands
Bridging Graph Drawing and Dimensionality Reduction with Stochastic Stress Optimization
Daniel Hangan¹, Stephen Kobourov¹, Jacob Miller¹
¹Technical University of Munich, Germany
Multivariate Graph Layouts using NNP-NET
Ilan Hartskeerl¹, Florian Grötschla², Alexandru Telea¹
¹Utrecht University, Netherlands · ²ETH Zurich, Switzerland
Topics
Examples of relevant topics for the workshop are as follows:
Connections between graph drawing and dimensionality reduction
Evaluation methodologies and interpretability
Cross-domain metrics, quality criteria, and scalability
Human-centered aspects of visualization (trust, stability, readability)
Comparative and methodological studies
Case studies and application scenarios combining GD and DR
Program Chairs
Alexandru Telea, Utrecht University, Netherlands
Alessandra Tappini, University of Perugia, Italy
Jean Ponciano, University of São Paulo, Brazil
Luis Gustavo Nonato, University of São Paulo, Brazil
Fabrizio Montecchiani, University of Perugia, Italy
Jacob Miller, Technical University of Munich, Germany
Lars Linsen, University of Münster, Germany
Hyeon Jeon, Seoul National University, Korea
Christophe Hurter, University of Toulouse, France
Thomas Höllt, TU Delft, Netherlands
Helwig Hauser, University of Bergen, Norway
Mohammad Ghoniem, Luxembourg Institute of Science and Technology, Luxembourg
Takanori Fujiwara, University of Arizona, United States
Velitchko Filipov, TU Wien, Austria
Michael Behrisch, Utrecht University, Netherlands
Important Dates
All deadlines are 23:59:59 Anywhere on Earth (GMT/UTC-12:00)
Submission Deadline: February 15, 2026 (Extended to February 28, 2026)
Notification of Acceptance: March 31, 2026 (Notifications sent!)
Camera-ready Papers Due: April 15, 2026
We are accepting submissions of 3–4 pages (including references) in two categories:
Short papers (4 pages): Present case studies, comparative analyses, or methodological contributions that highlight links between GD and DR. These may include empirical studies, critiques, or reports on cross-domain applications.
Position papers (3–4 pages): Offer conceptual arguments, speculative ideas, or reflective perspectives on methodological synergies between GD and DR.
All contributions should be submitted on PCS (https://new.precisionconference.com/submissions). We welcome both single-blind and double-blind submissions. Accepted contributions will be included in the Eurographics Digital Library. Selected papers will be invited to submit an extended version to a Special Issue on the Information Visualization Journal (https://journals.sagepub.com/home/IVI).
Link to the template and form for authors to prepare the camera-ready version: GDxDR Template, Copyright Form.
More details on the Call for Papers.
@Copyright 2026 Claudio D. G. Linhares