TPDL/ADBIS 2025 Program

Day 1 - Tuesday, 23rd September 2025
8.30-9.00
Conference Registration
9.00-11.00
11.00-11.30
Coffe Break
11.30-12.30
ADBIS/TPDL Keynote: Felix Naumann
12.30-14.00
Lunch Break
14.00-16.00
16.00-16.30
Coffe Break
16.30-17.30
Day 2 - Wednesday, 24th September 2025
8.00-8.40
Conference Registration
8.40-9.00
ADBIS & TPDL Opening
9.00-10.00
ADBIS/TPDL Keynote: Evaggelia Pitoura
10.00-10.30
Coffee Break
10.30-12.30
12.30-14.00
Lunch Break
14.00-16.00
16.00-16.30
Coffee Break
16.30-17.30
Panel Session - Do LLMs solve the Data Integration Problem once and for all?
18.00-19.00
Special Activity: Walking Tour
19.00-20.30
Special Activity: Visit to Old City Hall
Day 3 - Thursday, 25th September 2025
9.00-10.00
ADBIS/TPDL Keynote: Eero Hyvonen
10.00-10.30
Coffee Break
10.30-12.30
12.30-14.00
Lunch Break
14.00-16.00
16.00-16.30
Coffee Break
16.30-17.30
ADBIS/TPDL Demo Session
20.00-23.30
Dinner: LUMOAMO - Lapland Hotel & ADBIS/TPDL Best Paper Awards
Day 4 - Friday, 26th September 2025
9.00-10.00
ADBIS/TPDL Keynote: Yannis Ioannidis
10.00-10.30
Coffee Break
10.00-11.00
Poster Session
11.00-11.30
DEI Keynote: Maija Hirvonen
11.30-12.30
DEI Panel - Repositioning DEI: The Urgent Reimagining of DEI in an Era of Retrenchment
12.30-13.00
ADBIS & TPDL Closing Session

TPDL Session 1 - Large Language Models

  • Data Discovery Using LLMs - A Study of Data User Behaviour (FULL)
  • Generating Synthetic Data with Large Language Models for Low-Resource Sentence Retrieval (FULL)
  • Organising Knowledge from Text: Prompt-Based Triple Extraction and Graph Enrichment with Large Language Models (FULL)
  • LLM-Based Dependency Tracking for Short Event Descriptions (FINDINGS)
  • SciTableQA: A Question-Answering Benchmark for Complex Scientific Tables (FINDINGS)
Return to top

TPDL Session 2 - Scholarly Issues

  • From Raw Affiliations to Organization Identifiers (FULL)
  • How Retracted Article Persists on YouTube: Retraction Severity, Visibility, and Disclosure (FULL)
  • Neuro-symbolic Federated Research Artifact Search (FULL)
  • The Claims Network: Collecting Research, Education, Impact, and Leadership Claims on the Decentralized Web (FULL)
Return to top

TPDL Session 3 - Citation Management

  • Semantically Orthogonal Framework for Citation Classification: Disentangling Intent and Content (FULL)
  • Can LLMs Predict Citation Intent? An Experimental Analysis of In-context Learning and Fine-tuning on Open LLMs (FINDINGS)
  • Citation Impact and Research Persistence: The power of Positive Reinforcement (FINDINGS)
Return to top

TPDL Session 4 - Digital Archives

  • HistoryCLIP: Adaptive Multi-Modal Retrieval of Imbalanced Long-Tailed Archival Data (FULL)
  • Studying Biases in LLMs for Multilingual Historical Newspaper Extraction; The Messina Earthquake Case Study (FULL)
  • Human-Centred Design for Digital Ecosystems: Actors, Roles and Responsibilities (FINDINGS)
  • Real-time Prediction of Wikipedia Articles’ Quality (FINDINGS)
  • What Makes a User Click on a News Item? Understanding News Values of Visual Content in News Recommendation (FINDINGS)
Return to top

TPDL Session 5 - Digital Humanities and Cultural Heritage

  • Blowin' in the Wind: The Challenges of Archiving Social Media for Digital Humanities Research (FINDINGS)
  • Exploring Semantic and Motion-Based Documentation for Intangible Cultural Heritage Integration (FINDINGS)
  • The Text+ Registry: Federating Research Data Catalogues for the Digital Humanities (FINDINGS)
  • A Comparative Analysis of Support Techniques for Assessing the Quality of Systematic Literature Reviews (FINDINGS)
Return to top

TPDL Session 6 - Knowledge Graphs

  • An End-to-end Pipeline for Knowledge Graph Population from 19th-century Land Registry Digitised Tables (FULL)
  • EUR-Lex-Triples: A Legal Relation Extraction Dataset from European Legislation (FULL)
  • Heterogeneous Graph Representation for Dataset Link Prediction on Dynamic and Sparse Scholarly Graphs (FULL)
  • Human-In-The-Loop Workflow for Neuro-Symbolic Scholarly Knowledge Organization (FULL)
Return to top

ADBIS Session 1 - Query Optimization

  • Independence Rules: Analysis of Nine Simple Cardinality Estimators and Their Impact on Plan Quality (FULL)
  • Vector Database Benchmarking: A Face Retrieval Use Case (FULL)
  • PUL: Pre-load in Software for Caches Wouldn't Always Play Along (FULL)
  • iSearch: Seek acceleration through interpolation in Smart Storage Settings (SHORT)
  • Extending the Applicability of Bloom Filters by Relaxing their Parameter Constraints (SHORT)
Return to top

ADBIS Session 2 - Spatio-Temporal & Graph Data

  • Bitemporal Property Graphs: dealing with both valid and transaction time (FULL)
  • Could More be Less: The Case of Location(s) Awareness in Time Series Prediction (SHORT)
  • Spatio-Temporal Data and Molecular Dynamics: Challenges and Opportunities (Vision Paper) (SHORT)
  • From ER Conceptual Models to Document-based NoSQL Logical Models (SHORT)
  • Querying Property Graphs with XPath (SHORT)
Return to top

ADBIS Session 3 - Data Sharing & Synthesis

  • Generating Semantically Enriched Mobility Data from Travel Diaries (FULL)
  • Pasteur: Scaling Privacy-aware Data Synthesis (FULL)
  • Decentralized research data sharing management using blockchain technology (SHORT)
  • Experiversum: An Integrated Ecosystem for Curating and Enhancing Data-Driven Experimental Science (SHORT)
  • Extension of Data Catalog Vocabulary for federating open datasets in data spaces (SHORT)
Return to top

ADBIS Session 4 - Explainable AI

  • Responsible AI: Training deep learning model efficiently (FULL)
  • Exploring local feature influences with Hierarchical Explanation Trees (FULL)
  • Towards Trace Variant Explainability (FULL)
  • Evaluating Knowledge Generation and Self-Refinement Strategies for LLM-based Column Type Annotation (FULL)
Return to top

ADBIS Session 5 - Data & Machine Learning

  • Learning a Distance for the Stratification of Patients with Amyotrophic Lateral Sclerosis (FULL)
  • Designing Efficient and Scalable Substructure Discovery Algorithms for Multilayer Networks (FULL)
  • An Evaluation of Energy Consumption for Deep Learning-based Privacy Preserving Record Linkage (SHORT)
  • Leveraging Generative AI and Frequent Errors to Enhance SQL Learning (SHORT)
Return to top

ADBIS Session 6 - Entity Resolution & Integration

  • LSBlock: A Hybrid Blocking System Combining Lexical and Semantic Similarity Search for Record Linkage (FULL)
  • Evaluating Quality of Disparate Data Sources: A Discord-Driven Approach (FULL)
  • Accelerating Entity Resolution through Vectorized Meta-Blocking on GPUs (SHORT)
  • Validating Data Provenance Polynomials (SHORT)
  • Data Governance Maturity Models: A Systematic Review and a Multi-Dimensional Framework Proposal (SHORT)
Return to top

Guidelines

  • FULL papers: 20mins presentation + 6mins Q/A
  • SHORT papers: 15mins presentation + 5mins Q/A
  • FINDINGS papers: 15mins presentation + 5mins Q/A