These are the proceedings of the Second International Workshop on Scholarly Information Access (SCOLIA 2026)1. SCOLIA 2026 was held as a half-day in-person event at the European Conference on Information Retrieval (ECIR) in Delft, The Netherlands. The aim of the SCOLIA workshop, following the successful and long-established BIR workshop series [1, 2], is to bring together researchers and practitioners from Information Retrieval (IR), Natural Language Processing (NLP), and Scientometrics/Bibliometrics who are working on the analysis of scientific/scholarly documents. Bibliometrics and Scientometrics are concerned with all quantitative aspects of information and academic literature, which naturally make them interesting for IR research, as well as domains in which citations play a central role, for example, legal and patent retrieval. The SCOLIA workshop has a broader scope than the previous BIR workshop series. This allows us to cater for recent developments in NLP and Generative AI, which are becoming more important for SCOLIA's core focus on scholarly information access and recommendation. The workshop received 17 submissions across three tracks: full papers, short papers and shared task proposals. The following eight papers (three full, four short and one shared task) papers papers are included in the proceedings. All articles were peer-reviewed by at least two experts in the field. The works were presented in three sessions during the workshop.
embedding models
Full paper • Chun-Hsi Ku and Hung-Hsuan Chen:
Structure-Preserving Graph Contrastive Learning for Mathematical Information Retrieval Short paper • Aditya Neekhra, Markus Nilles and Ralf Schenkel:
RAGScholar & DBLP-QA: Explainable Retrieval Augmented Scientific QA on dblp with SourceAttributed Answers and a Benchmark Dataset
Short paper Session 3: Systematic Reviews • Pierre Achkar, Tim Gollub and Martin Potthast:
A Test Collection of Systematic Reviews in Computer Science
Short paper • Pierre Achkar, Tim Gollub, Martin Potthast, Carsten Eickhof and Harrisen Scells: UP2DATE: Shared Task on Systematic Review Updates
Shared task
We thank our reviewers.
• Akiko Aizawa; National Institute of Informatics, Japan • Patrice Bellot; Aix-Marseille Université - CNRS (LIS), France • Marc Bertin; Université Claude Bernard Lyon 1, France • Katriina Byström; Oslo Metropolitan University, Norway • Xingyu Deng; University of Shefield, UK • Norbert Fuhr; University of Duisburg-Essen, Germany • Tim Gollub; Bauhaus-Universität Weimar, Germany • Andreas Henrich; University of Bamberg, Germany • Roman Kern; Know-Center GmbH, Austria • Hermann Kroll; TU Braunschweig, Germany • Udo Kruschwitz; University of Regensburg, Germany • Cyril Labbé; Grenoble Alpes University, France • Anastasios Nentidis; NSCR Demokritos, Greece • Markus Nilles; Trier University, Germany • Rajesh Piryani; IRIT - Equipe MELODI Universite Paul Sabatier, France • Martin Potthast; University of Kassel, Germany • Florian Reitz; dblp, Germany • Libo Ren; Modul University Vienna, Austria • Thomas Risse; University Frankfurt, University Library J. C. Senckenberg, Germany • Harald Sack; FIZ Karlsruhe & KIT Karlsruhe, Germany • Philipp Schaer; TH Köln -University of Applied Sciences, Germany • Ralf Schenkel; Trier University, Germany • Mark Stevenson; University of Shefield, UK • Xi Wang; University of Shefield, UK • Dietmar Wolfram; University of Wisconsin-Milwaukee, USA This work was funded by the European Union under the Horizon Europe grant OMINO – Overcoming Multilevel INformation Overload (grant number 101086321, http://ominoproject.eu).
The authors have not employed any Generative AI tools.
The SCOLIA 2026 Organisers