The rapid growth of online available scientific, technical, and legal data such as patents, reports, articles, etc. has made the large-scale analysis and processing of such data a crucial task. Today, scientists, patent experts, inventors, and other information professionals (e.g., information scientists, lawyers, etc.) contribute to this data every day by publishing articles, writing technical reports, or patent applications.
It is a challenging task to process, analyze, and explore documents due to their length, the use of domain-specific vocabulary, and the complexity introduced by targeting various scientific fields and domains. Documents are semi-structured and cover unstructured textual parts as well as structured parts such as tables, mathematical formulas, diagrams, and domain-specific information such as chemical names, bio-sequences, etc.
Such kind of information brings complexity in processing such documents; however, data is the lifeblood of many applications, and its preservation, analysis, enrichment, and use are key for applications in several domains. In order to benefit from the scientific-technical knowledge present in such documents, e.g., for decision-making or for professional search and analytics, there is an urgent need for analyzing, enriching, and linking such data by employing state-of-the-art Semantic Web technologies and AI methods.
However, as they are heterogeneous and are written using domain-specific terminology applying the existing semantic technologies is not straightforward. To address the challenges mentioned above, Semantic Web Technologies, Natural Language Processing (NLP) techniques, Deep Neural Networks (DNN), and Large Language Models (LLMs) must be leveraged in order to provide efficient and effective solutions for creating easily accessible and machine-understandable knowledge.
Contact us if you did not make it on time!
The workshop accepts contributions in all topics related to semantic web technologies and deep learning focused (but not limited) to:
The submissions must be in English and adhere to the CEUR-WS one-column template (see Session 2: The New CEURART Style). The papers should be submitted as PDF files to EasyChair. The review process will be single-blind. Please be aware that at least one author per paper must be registered and attend the workshop to present the work and that ESWC is a 100% in person conference.
We will consider three different submission types:
Submissions should not exceed the indicated number of pages, including any diagrams and references.
Each submission will be reviewed by three independent reviewers on the basis of relevance for the workshop, novelty/originality, significance, technical quality and correctness, quality and clarity of presentation, quality of references and reproducibility.
The accepted papers will be available on the Workshop website. The proceedings will be published in a CEUR-WS volume and consequently indexed on Google Scholar, DBLP, and Scopus.
All the information to register and attend the workshop can be found on the ESWC registration page.
SemTech4STLD workshop will take place on June 1st, 2025.
| Timing | Content |
|---|---|
| 9:00 9:10 |
Opening & Welcome
|
| 9:10 10:00 |
Keynote and Q&A on Evaluation Challenges in Using Generative AI for Science & Technical Content
Speaker: Prof. Dr. Paul Groth INDE Lab, University of Amsterdam Abstract: Foundation Models show impressive results in a wide-range of tasks on scientific and legal content from information extraction to question answering and even literature synthesis. However, standard evaluation approaches (e.g. comparing to ground truth) often do not seem to work. Qualitatively the results look great but quantitive scores do not align with these observations. In this talk, I discuss the challenges we have faced in our lab in evaluation. I then outline potential routes forward. Short Bio: Paul Groth is Professor of Algorithmic Data Science at the University of Amsterdam where he leads the Intelligent Data Engineering Lab (INDElab). He holds a Ph.D. in Computer Science from the University of Southampton (2007) and has done research at the University of Southern California, the Vrije Universiteit Amsterdam and Elsevier Labs. His research focuses on intelligent systems for dealing with large amounts of diverse contextualized knowledge with a particular focus on web and science applications. This includes research in data provenance, data integration and knowledge sharing. Paul is scientific director of the UvA’s Data Science Center. Additionally, he is co-scientific director of two Innovation Center for Artificial Intelligence (ICAI) labs: The AI for Retail (AIR) Lab - a collaboration between UvA and Ahold Delhaize; and the Discovery Lab - a collaboration between Elsevier, the University of Amsterdam and VU University Amsterdam. Previously, Paul led the design of a number of large scale data integration and knowledge graph construction efforts in the biomedical domain. Paul was co-chair of the W3C Provenance Working Group that created a standard for provenance interchange. He has also contributed to the emergence of community initiatives to build a better scholarly ecosystem including altmetrics and the FAIR data principles. Paul is co-author of “Provenance: an Introduction to PROV” and “The Semantic Web Primer: 3rd Edition” as well as numerous academic articles. You can find him on twitter: @pgroth . |
| 10:00 10:15 Session I |
Paper I: Enabling Natural Language Access to BIM Models with AI and Knowledge Graphs Andrea Ibba, Ruben Alonso and Diego Reforgiato Recupero, (12 min + 5 Q&A)
|
| 10:30 11:00 |
Coffee Break
|
| 11:00 12:25 Session II |
Paper II: Biomedical Entity Linking with Triple-aware Pre-Training, Xi Yan, Cedric Möller and Ricardo Usbeck, (12 min + 5 Q&A)
Paper III: Evaluating LLMs for Named Entity Recognition in Scientific Domain with Fine-Tuning and Few-Shot LearningDavide Buscaldi, Danilo Dessì, Francesco Osborne, Davide Piras and Diego Reforgiato Recupero, (16 min + 5 Q&A)
Paper IV: Benchmarking Large Language Models for Sustainable Development Goals Classification: Evaluating In-Context Learning and Fine-Tuning StrategiesAndrea Cadeddu, Alessandro Chessa, Vincenzo De Leo, Gianni Fenu, Enrico Motta, Francesco Osborne, Diego Reforgiato Recupero, Angelo Salatino and Luca Secchi, (12 min + 5 Q&A)
Paper V: Taming Hallucinations: A Semantic Matching Evaluation Framework for LLM-Generated OntologiesNadeen Fathallah, Steffen Staab and Alsayed Algergawy, (16 min + 5 Q&A)
Paper V & VI (Posters):
Leveraging Knowledge Graphs and Generative AI for Augmented Research Paper RetrievalRima Dessi', Erick Mendez Guzman, Alya Alshaami, Amna Alowais, Hamda Alhammadi, Nada Alzarooni, Weam N.A Jarbou and Zarak Khan
Context-Aware Explanations: Leveraging Knowledge Graphs for Adaptive Explainability in Dynamic EnvironmentsErick Mendez Guzman, Rima Dessi', Alya Alshaami, Amna Alowais, Hamda Alhammadi, Nada Alzarooni, Weam N.A Jarbou and Zarak Khan |
| 12:25 12:30 |
Closing Remarks
|