13:30 : Welcome and Introduction.
13:30 - 14:15 : Keynote by Luc Moreau
Title : Unlocking the Power of Provenance with Templates
Abstract: In today’s data-driven world, understanding where data comes from—and how it is used—is crucial for fostering trust, transparency, and accountability. Provenance, which documents the origins and transformations of data, is essential to this understanding. However, developing provenance-enabled applications—namely, applications that create, manage, and utilise provenance—can be both cumbersome and time-consuming. This is where provenance templates play a significant role. Provenance templates are a powerful approach that streamlines the creation of provenance through reusable and processable structures. Consider them as blueprints with placeholders that can be automatically filled in and customised during data processing. Once instantiated, these templates produce PROV documents—a structured and standardised representation of provenance—ready for sharing and analysis. An additional feature of provenance templates is that they are themselves provenance documents, and the instantiation of a provenance template can lead to the creation of another provenance template, opening up a range of opportunities to manage provenance templates systematically. I further introduce the “Provenance First” approach, which specifies a contract between the components that produce provenance and those that utilise it through provenance templates. This enables us to integrate provenance design into the software development lifecycle and, with the appropriate tooling, facilitate the development of a provenance-enabled application. At the heart of this innovation lies the Provenance Template Compiler. This tool generates ready-to-use libraries that seamlessly integrate with applications. The result? Provenance is generated automatically and consistently, making it easy to embed into existing systems. This work reimagines how provenance is created and managed, transforming it into a practical, scalable asset for organisations in science, business, government, and beyond. Regardless of the application being developed, provenance templates provide clarity, confidence, and control to your data journey. However, the benefits extend beyond this. By employing templates, organisations can facilitate advanced features such as provenance access control, provenance integrity verification, provenance visualisations, and even automated explanations. Ultimately, this approach addresses a significant barrier to widespread provenance adoption: the lack of practical tools and awareness, rather than technological limitations.
Bio : Luc Moreau is a Professor of Computer Science. He joined the University of Sussex, UK, as Executive Dean of the Faculty of Science, Engineering and Medicine in September 2024. Before his arrival at Sussex, Luc served as the Head of the Department of Informatics at King's College London, where he led a strategic expansion of the department, enhancing its scope and capabilities. Prior to joining King's, Luc was the Head of Web and Internet Science in the Department of Electronics and Computer Science at the University of Southampton. Luc served as co-chair of the Provenance Working Group at the World Wide Web Consortium (W3C), which led to the development of four W3C Recommendations and nine W3C Notes specifying PROV, a conceptual data model for provenance on the Web, along with its serialisations in various Web languages. Previously, he initiated the successful Provenance Challenge series, involving over 20 institutions in investigating provenance interoperability across three successive challenges, ultimately resulting in the specification of the community Open Provenance Model (OPM). He co-founded the workshop series IPAW and Provenance Week, which have been flagship venues for the provenance community for nearly two decades.
14:15 - 14:30 : Abdullah Almuntashiri, Luis-Daniel Ibáñez, and Adriane Chapman, Using LLMs to infer provenance information
14:30 - 14:45 : Xiaoyu Han, Haneen Mohammed, Charlie Summers, and Eugene Wu, Compression for High-Performance Lineage
14:45 - 15:00 : Antoine Groudiev, Arkaprava Saha, and Silviu Maniu, Extending Layer-wise Relevance Propagation in Neural Networks using Semiring Annotations
15:00 - 15:30 : Coffee Break
Chair : TBD
15:30 - 15:45 : Bernard Roper and Heather Packer, PROVPub: A Publication Model to Support Research Evaluation and Acknowledgement
15:45 - 16:00 : Bertram Ludaescher, Yilin Xia, and Shawn Bowers, Choices and their Provenance: Explaining Stable Solutions of Abstract Argumentation Frameworks
16:00 - 16:15 : Pitch (Lightning) Talks
16:15 - 17:00 : Poster Session
17:00 : Closing