The conference will be held live on campus on July 3rd 2026.
The event is divided into sessions. There will be a plenary session in the morning where everyone should be present; afterwards the different track sessions will happen in parallel. The plenary session contains the pitches of the best papers nominated from all the submissions, followed by an award ceremony where the very best will be awarded.
Notes on the presentations:
The slots allocated for the presentations at the conference will be 15 minutes per paper. This includes the presentation itself (min. 10 and max. 12 minutes), and 3 minutes for questions. There will be not format for the slides. Students are free to select their preferred format, as long as it is clean and professional. All presentations will be live on campus (apart from exceptional cases).
The proceedings will be published online with open access policies. The proceedings of the Past editions are still available for your reference.
There will be two main sessions:
The plenary session (8:45 – 9:15) in Ravelijn 1501
Parallel track sessions (9:30 – 15:45, depending on tracks) in Ravelijn
Drinks at bordes at the entrance of Zilverling (starting at 14:30, depending on tracks)
Note: The presentation order within the track sessions is organized by the track chair and mainly depending on the availability of the supervisors. If you have special requests, please contact your track chair.
8:45 Opening
8:55 Introduction of Best Paper nominations
9:00 Award ceremony Best Paper
9:15 Break (coffee/tea in Ravelijn)
9:30 Track welcomes
9:45 Parallel track sessions (Ravelijn)
14:30 Drinks at bordes at the entrance of Zilverling
Track 1
Matthijs Jansen Op de Haar - GuidedRAG: Semantic Steering of Retrieval Augmented Generation
Ernests Malnacs - Motion Sickness Through Thermal Cue Incongruence
Track 2
Cristian Babin - Response-Based LLM Route Switching for Financial Regulatory Compliance: Towards Cost-Efficient and Accountable AI Deployment
Ervinas Vilkaitis - Controlling LLM Responses with Clinical Rules: A Hybrid Rule-Based Escalation Architecture for IVF Patient Query Triage
Track 3
Nojus Morkūnas - Integrating Model-Based Testing into Agile Software Engineering Workflows
Vlad Negară - SCSS without the S: Automated refactorings for maintainability using modern CSS features
Track 4
Saksham Singh Birla - Evaluating Multimodal Fusion Robustness in Drone-Based Object Detection Under Sensor Degradation
Track 5
Tristan Bottenberg - DeltaParq - Using Delta Compression for Efficient Storage of OpenINTEL’s Historical DNS Records
Track 6
Diego Arroyo Socorro - How Sparsity and Training Length Shape Pure PPO Survival on a Power-Grid Benchmark
Danil Badarev - Pruning as a Fine-Tuning Paradigm for Large Language Models: Evaluating Sparsity Evolution in Persona Generation
Track 7
Daniel Botezatu - Autonomous LLM-Driven RTL Generation for Deep Neural Networks via Multi-Agent Orchestration
Junseo Kim - Combining Recursive Weight-Sharing with Token Merging for Edge Vision Transformers
Track 8
Tudor Matei - Request Cancellation in Web Application Frameworks
Steef Broeder - Investigating the safety/security interactions of the Dutch transmission grid using Attack-Fault-Defense Trees
Track chair: Mariet Theune
RA 2231
09:45 – 10:00 - Alexandru-Cristian Enescu - Generalization and Adaptability of ACT-based Imitation Learning and Interactive Imitation Learning in Cobot Pick-and-Place Tasks under Spatial Distribution Shifts along a Single Axis
10:00 – 10:15 - Masha Lindenaar - Design of a monitoring dashboard for a modular conveyor system
10:15 – 10:30 - Matthijs Jansen Op de Haar - GuidedRAG: Semantic Steering of Retrieval Augmented Generation
10:30 – 10:45 - Maja Korzeniowska-Jęczeń - Emotional Support LLM Chatbot: Enhancing the Therapeutic Alliance through Adaptive Communication Styles.
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Ou-Xiang Chen - Flexible Interactive Tablecloth
11:15 – 11:30 - Carlo Fernandes de Brito - Comparing Time-Based and Context-Aware Smartwatch Prompting: Effects on Participant Experience in ESM Studies
11:30 – 11:45 - Hugo Dijkema - Capturing university students’ sense of belonging through ESM, passive sensing, and stimulated recall: a mixed-methods study
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Robert Pit - Developing an e-learning module for inclusive software design
13:30 – 13:45 - Ivan Andonov - Pressure-sensitive inner sole for gait analysis
13:45 – 14:00 - Shashank Bajoria - Detecting Doomscrolling: A Passive Smartphone Sensing Approach
14:00 – 14:15 - Suraj Mishra - Into the Habit Loop: Detecting Compulsive Smartphone Checking Behaviour Through Passive Sensor Data
14:15 – 14:30 Minibreak
14:30 – 14:45 - Milan Chapkin - Warning Labels, Verified Humans, and Engagement Gates: Testing Interface Cues in Synthetic Social Media Comment Sections
14:45 – 15:00 - Daria Hesson - Understanding Social Media Challenges between Parents and their Children
15:00 – 15:15 - Tara van Haarst - Detecting Deceptive Design in Children's Mobile Applications: An Ontology-Based Analysis and Rule-Based Prototype Tool
15:15 – 15:30 Poll for Best Presentation
RA 2237
09:45 – 10:00 - Ilan Westerhof - Simulation of the Professional Motorcycle Experience
10:00 – 10:15 - Ernests Malnacs - Motion Sickness Through Thermal Cue Incongruence
10:15 – 10:30 - Harm Dreteler - Comparing Effectiveness of Robot Ball Behaviour Patterns at stimulating Physical Activity through Play in Physical Therapy
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Andreea Serban - Redesigning the Social VR Museum Experience: Evaluating Collaborative versus Competitive Game Mechanics
11:15 – 11:30 - Sanne Lohuis - How non-Euclidean environments affect attention span to and memory retention of art in virtual museum
11:30 – 11:45 - Bogdan Mocanu - Designing and Evaluating a Context-Aware Virtual Reality Onboarding Experience for Museum Visitors
11:45 – 12:00 - Andrei Rusu - ML vs CNN Approach for Stress Detection using Time Series Data
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Mirza Johanningmeijer - Evaluating the Robustness of Time Series Analysis Techniques for Stress Detection Using Wearable Physiological Data
13:30 – 13:45 - Begaiym Sarbagysheva - Design Requirements for a Conversational Agent Interface to Support Clinical Assessment for Individuals with Mild to Borderline Intellectual Disability Undergoing Addiction Care
13:45 – 14:00 - Sabo Fidan - Design Recommendations for Conversational Agents to Support Clinical Assessments in Patients with Substance Use Disorder.
14:00 – 14:15 - Simay Odabaşı - Designing A Planning Tool for University Study Advisers to Enhance Neuro-Inclusive Education
14:15 – 14:30 Poll for Best Presentation
Track chair: Marcos Machado
RA 2501
09:45 – 10:00 - Christina Braun - Optimizing Retrieval Quality in Reproductive Health RAG Systems: A Comparative Study of Chunking Strategies and Embedding Models
10:00 – 10:15 - Tania Maria Mincu - Designing an Extension to BuddyGPT: A RAG-Based AI Chatbot for Sustainable Nutrition Guidance for University Students
10:15 – 10:30 - Ioana Natasa Tudorache - Designing and Evaluating a RAG-Based Conversational AI for Supporting Sustainable Nutrition Consumption Behaviour in Female Student Population
10:30 – 10:45 - Ervinas Vilkaitis - Controlling LLM Responses with Clinical Rules: A Hybrid Rule-Based Escalation Architecture for IVF Patient Query Triage
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Levi Compen - Bridging the Gap between Alcohol-Free and Regular Beer: an ML-driven Advisory Framework
11:15 – 11:30 - Sabin Mihai Nicolae - Ethical Concerns Surrounding Consumer Smart Doorbells: A Grounded Literature Review with User Perspectives
11:30 – 11:45 - Mustafa Sen - Trust in AI Chatbots: University Students’ Perceptions of Confident and Uncertainty-Expressing Responses
11:45 – 12:00 - Ugnius Tulaba - Benchmarking Quantization and Pruning on ResNet-18 for CIFAR-10 Image Classification
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Cristian Babin - Response-Based LLM Route Switching for Financial Regulatory Compliance: Towards Cost-Efficient and Accountable AI Deployment
13:30 – 13:45 - Tieme van Enkhuizen - Study of Single-Agent and Multi-Agent LLM Systems for Document Recognition and Data Mapping in SMEs
13:45 – 14:00 - Marius Inselseth - Do Planner-Executor Agents Justify Their Cost in Constraint-Heavy Travel Planning?
14:00 – 14:15 - Giorgos Kyriakou - How Stable Are Few-Shot Prompting Trade-Offs? A Cross-Task Cost-Utility Analysis for LLM Text Classification
14:15 – 14:30 Minibreak
14:30 – 14:45 - Kristian Lenkmann - Prompt Compression on Commercial APIs: Evaluating Net Cost and Output Quality Across Task Types
14:45 – 15:00 - Diego Moreno de Miguel - Task-Based Model Switching for Controllable Cost-Quality Trade-offs in IT Service Management
15:00 – 15:15 - Mara Bianca Teodorescu - The Cost of Oversizing: Quantifying the Cost-Quality Trade-off of Model Right-Sizing in Document Summarisation
15:15 – 15:30 - Lenn Beld - Governing the risk of AI reliance: engineering capability retention in software companies.
15:30 – 15:45 Poll for Best Presentation
RA 2502
10:00 – 10:15 - Duc Cuong Bui - A Global Overview of Anti-Fraud Awareness Campaigns: Effectiveness and Implications for Financial Authorities.
10:15 – 10:30 - Jesse van Dijk - From Noise to Narrative: A visualization tool for DDoS events across PEST dimension
10:30 – 10:45 - Omar Alaaeldin Badreldin Moustafa - A (Digital) Networking Game: Designing and Testing a Hybrid Game That Teaches How the Internet Works
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Harshit Swarna - An Office-based Task For Phishing Simulation Experiment
11:15 – 11:30 - Daria Zvinca - Assessing the Effectiveness of Countermeasures Against Social Engineering Attacks: A Meta-Analysis
11:30 – 11:45 - Emil Muradli - To Shoot Or Not To Shoot: Analyzing Players’ Shooting Inhibition in Counter-Strike 2
11:45 – 12:00 - Pramudito Nirwikara - The Effects of Ambiguous Stimuli on Response Inhibition of CounterStrike 2 Players
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Alexia-Georgia Opinca - AI Applications in Crowdfunding: Use, Access, and Implications for Social Inclusion
13:30 – 13:45 - Ezgi Ozenc - Artificial Intelligence Applications in Crowdfunding: A Systematic Review of Applications and Links to Sustainable Development Goals
13:45 – 14:00 - Etibar Aliyev - Investment Dynamics and Industrial Symbiosis Network Position in the Basque Hub for Circularity
14:00 – 14:15 - Cosmin Ana - Is the metaverse dead? A Comprehensive Analysis of Metaverse Related Companies and Projects
14:15 – 14:30 Minibreak
14:30 – 14:45 - Oguz Kaan Aydin - A Multi-Dimensional Analysis of Cybersecurity Incident Severity Across Organizations
14:45 – 15:00 - Jiya Kartoidjojo - A Framework of Cryptographic Asset Discovery and Inventory (CADI) for IoT, OT, and ICS Systems
15:00 – 15:15 - Emirhan Kitapci - From "Human Error" to Systemic Analysis: Utilizing Attack Patterns for Organizational Learning in Cybersecurity.
15:15 – 15:30 - Anna Hovenkamp - The Impact of Quishing: A Safety and Security Interaction Model Using the BDMP Formalism
15:30 – 15:45 - Ruslan Kasac - Mitigating QR Phishing Through Combined QR code and URL Analysis
15:45 – 16:00 Poll for Best Presentation
Track chair: Moritz Hahn
RA 2503
09:45 – 10:00 - Floortje ter Avest - The Relation Between Modern Code Practices and Code Quality
10:00 – 10:15 - Michal Dac - Me Prompt, You Code: Prompt Verbosity and Token Efficiency in LLM-Assisted TDD
10:00 – 10:15 - Maarten van Dort - Feature Adoption in Golang: An Empirical Study of Post-Generics Go Evolution
10:15 – 10:30 - Evelyn van der Sar - Measuring codebase modernity in Kotlin
10:30 – 10:45 - Wiktor Brach - What Drives Babbling? The Role of Type Systems and Model Scale in LLM Code Generation
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Giulia Tălău - How Modern are Java Codebases?
11:15 – 11:30 - Lars Wijntjes - Codebase Modernity in Lua
11:30 – 11:45 - Michal Dac - Me Prompt, You Code: Prompt Verbosity and Token Efficiency in LLM-Assisted TDD
11:45 – 12:00 - Vlad Negară - SCSS without the S: Automated refactorings for maintainability using modern CSS features
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Alisa Biclesanu - High-Level Modeling of CTMC Queueing Systems for JANI-Compatible Probabilistic Model Checking
13:30 – 13:45 - József Márton Kakas - Modeling and Verification of Keeta’s Two-Phase Consensus Protocol
14:00 – 14:15 - Milan Oosterink - Autonomous Reward Engineering for 2D Vehicle Control: An Evolutionary Approach Using Large Language Models
14:15 – 14:30 Minibreak
14:30 – 14:45 - Ivan Peñate Gómez - A JANI-Exporting Graphical Editor for 3D Grid-Worlds
14:45 – 15:00 - Iris Borgonjen - Gamification of Deductive Systems for Propositional Logic
15:00 – 15:15 - Alexandra Jakovleva - Extending Smoke Tests in VerCors for Detecting Specification Inconsistencies
15:15 – 15:30 Poll for Best Presentation
RA 2504
09:45 – 10:00 - Melle Bosboom - Compiling a Minimal Lazily Evaluated Language
10:00 – 10:15 - Twan Kuipers - Compiler-Driven Optimisation of First-Class Behavioral Contracts via LLVM IR Metadata
10:15 – 10:30 - Niek Peters - Bridging the Gap between Mathematics and Programming via a Dual-Target DSL for LaTeX and Haskell
10:30 – 10:45 - Lucas Pruijssers - Introducing Density: A Programming Language Designed for Procedural Terrain Generation
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Mart Dikker - Improving Decision Tree to Fault Tree Conversion
11:15 – 11:30 - Nojus Morkūnas - Integrating Model-Based Testing into Agile Software Engineering Workflows
11:30 – 11:45 - Lars van Soest - Exploring software fault tree generation in their approach, language paradigms and limitations
11:45 – 12:00 - Ryan Cooijmans - Modelling and analysing Yahtzee as a Markov Decision Process in Storm
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Mohamed Mohamedin - Smaller Synthesised Circuits by Post-Processing Parity Game Strategies through Edge Removal
13:30 – 13:45 - Adam Tozser - Fixing the Factorio Rail Planner: Formal Analysis of a Real-World Constrained Pathfinding Problem
13:45 – 14:00 - Stefan Veltmaat - Can Lace.rs Improve OxiDD’s Parallel Performance?
14:00 – 14:15 - Furqan Saleh - Investigating the notion of Evidence in CTL*
14:15 – 14:30 Poll for Best Presentation
Track chair: Duc le Viet
RA 2334
09:45 – 10:00 - Saksham Singh Birla - Evaluating Multimodal Fusion Robustness in Drone-Based Object Detection Under Sensor Degradation
10:00 – 10:15 - Deividas Buinevicius - Drone-Based Object Detection and Explanation Using Vision-Language Models
10:15 – 10:30 - Deniss Korņijenko - Implementation of a High-Availability LoRaMesh Network for Resilient Firefighter Communications on the Raspberry Pi Ecosystem
10:45 – 11:00 Coffee Break
11:15 – 11:30 - Stan Groote Stroek - Classifying Waste Materials from Multispectral Satellite Imagery
11:30 – 11:45 - Koppány Heizer - Detecting Trees from Satellite Imagery
11:45 – 12:00 - Lyubomir Yosifov - Detecting Swimming Pools from Satellite Imagery Employing Multiple Modalities
12:00 – 13:15 Lunch Break
13:15 – 13:30 - David-Filip Ionas - Analyzing the Impact of Latency on Collaborative vs Competitive AR Games
13:30 – 13:45 - Boaz Voort - Evaluating Mode Switching Strategies for Gaze-and-Foot Interaction in VR Object Selection and Manipulation
13:45 – 14:00 - Max van Zanten - Evaluating the Class-wise Impact of Model Compression in RF-Based Human Activity Recognition
14:00 – 14:15 - Artur Mucowski - LLM-Based Conflict Resolution Between Monitoring Agents in a Simulated Cyber-Physical Factory
14:15 – 14:30 Minibreak
14:30 – 14:45 - Ewout van Dijk - Optimizing the connectivity within a maritime MANET using a drone-swarm
14:45 – 15:00 - Jibbe Konniger - GridLock: Cooperative Localisation in Wireless Mesh Networks using Gaussian Belief Propagation
15:00 – 15:15 - Taylan Kıncır - Continual Reinforcement Learning for Robotic Navigation: Benchmark Design and Comparative Evaluation of CRL Methods
15:15 – 15:30 - Stefan Munteanu - CleanTools: A Unified Framework for Evaluating Tool-Misuse Defenses in LLM Agents
15:30 – 15:45 - Marcus de Lange - Detecting a Quantum Die has been hit against another using IMU Data
15:45 – 16:00 Poll for Best Presentation
Track chair: Suzan Bayhan
RA 3231
09:45 – 10:00 - Ryan Bartelds - Digging in DNS history: Designing a DNS name-server for querying historical OpenINTEL data
10:00 – 10:15 - Tristan Bottenberg - DeltaParq - Using Delta Compression for Efficient Storage of OpenINTEL’s Historical DNS Records
10:15 – 10:30 - Radu Roznovat - Life of a domain name - Episode 2: email
10:30 – 10:45 - Kevin Veldman - Tracking Network Infrastructure Dependency Changes in ccTLDs
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Konstantinos Kokkalakis - AI-based Decision Making for Edge Resource Management
11:15 – 11:30 - Maria Antonia Petrea - Post-Quantum Cryptography Impact on Network Bandwidth in SAML-Based Federated Identity Management Architectures
11:30 – 11:45 - Tim van Beek - Evaluating National Transitions to Post-Quantum Cryptography: A Framework-based Approach with a Case Study of South Korea
11:45 – 12:00 - Sami Elmusharaf - Evaluating the Computational Load of Post-Quantum Cryptography in Federated Identity Management Architectures Using OpenID Connect
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Tim Hendriks - Analysing Interaction Paths in a Hate-Labelled Reddit Reply Network Using a Multi-Constraint Approach
13:30 – 13:45 - Daniel Murse-Caraian - SEIZ-aware gravity centrality for ranking influential spreaders in Reddit interaction networks
13:45 – 14:00 - Ashkan Nikjouyan - Stability of Centrality Based Influential Users Rankings in Reddit Interaction Networks under Network Modification
14:00 – 14:15 - Alexandru Stratan - HateScope: An Interactive Multi-Platform Dashboard for Analysing Hate Speech Propagation Across Social Media
14:15 – 14:30 Minibreak
14:30 – 14:45 - Konstantinos Zygouris - Measuring the Temporal Stability of User Influence in Reddit-Based Social Networks
14:45 – 15:00 - Shounak Malekar - Failure Analysis of Multimodal Language Model in Detecting Scam Like Content in Short-Form Social Media Videos
15:00 – 15:15 Poll for Best Presentation
RA 3237
10:00 – 10:15 - Arne Pilz - Impact of Carbon-Aware Service Endpoint Selection on Network Paths
10:15 – 10:30 - Raul Botea - AS0 ROAs in the Wild: Understanding Intentional Route Suppression in the RPKI
10:30 – 10:45 - Maia Souvannasouck - Measuring the Observability of AWS Sovereign Cloud in Europe
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Wessel Ritskes - Inferring Router-Level and AS-Level Internet Topology from BGP Scan Data
11:15 – 11:30 - Rune Ebbers - Deploying QUIC Using Anycast
11:30 – 11:45 - Yi Hu - Home or Away? Measuring Anycast Replica Locality in EU ccTLD Authoritative DNS
11:45 – 12:00 - Victor Burcovschi - A Composite Placement Model for Climate-Neutral Edge Data Centers in the Netherlands
12:00 – 13:15 Lunch Break
13:30 – 13:45 - Marijn de Haan - Multilingual LLM Website Classification: The Effects of Language Resource Level, Script Type, and Input Representation
13:45 – 14:00 - Estere Saliņa - Trust in Numbers: A Study of IP Address Certification Trends in TLS Certificates
14:00 – 14:15 - Marten Aan de Stegge - The Impact of Free, Automated IP Certificates: Analyzing Let's Encrypt's Effect on Volume, Address Spaces, and Migration Patterns
14:30 – 14:45 Poll for Best Presentation
Track chair: Kostas Georgiadis
RA 4334
09:45 – 10:00 - Anna Gololobova - Foundation Models for Robust Medical Image Analysis: Detecting Unseen Histology Samples
10:00 – 10:15 - Roland Garvasuc - Zero-Shot and Fine-Tuned Foundation Models for Polyp Segmentation: A Comparative Study of SAM and MedSAM on the Kvasir-SEG Dataset
10:15 – 10:30 - Sumi Jang - Explainable Artificial Intelligence for Medical AI Models: A Case Study on Chest X-ray Classification
10:30 – 10:45 - Yash Sakore - Zero-Shot Food Recognition Across Cultures: Prompt Engineering and Fine-Tuning of CLIP for Non-Western Food Classification
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Arseniy Tokarev - Face Detection in the Practical Conditions: Evaluating MTCNN Performance, Transfer Learning, and Hardware Efficiency on the WiderFace Dataset
11:15 – 11:30 - Bekjan Barchynbekov - Face Recognition Performance on CPU vs GPU
11:30 – 11:45 - Tudor Calin - Text-Prompted Lung Segmentation for AI-Assisted Diagnosis: A Comparative Evaluation of BiomedCLIP and MedSAM
11:45 – 12:00 - Sonia Tofeni - Using Large Language Models to Support the Interpretation of Subpopulation Process Comparison Techniques in Process Mining
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Teodor Constantinescu - Development of image registration methods for art historical collections
13:30 – 13:45 - Lara Naz Hashas - Deep Learning-Based Image Segmentation of Cultural Heritage Objects
13:45 – 14:00 - Alexandra Lupan - Automatic Feature Extraction from Fingerprint Images on Cultural Heritage Objects
14:00 – 14:15 - Reinder Grondsma - Unsupervised Integration of Multi-Index Satellite Time-Series Features for Characterising Potato Growth Cycles
14:15 – 14:30 Minibreak
14:30 – 14:45 - Diego Arroyo Socorro - How Sparsity and Training Length Shape Pure PPO Survival on a Power-Grid Benchmark
14:45 – 15:00 - David Szentpeteri - Evaluating Sparse Neural Networks in Neural SARSA and DQN for Stock Trading
15:00 – 15:15 - Danil Badarev - Pruning as a Fine-Tuning Paradigm for Large Language Models: Evaluating Sparsity Evolution in Persona Generation
15:15 – 15:30 - Eren Atakan - Beyond Aggregate Statistics: A Per-Rally Feature-Importance Analysis of Tactical Patterns in the Sony Ace Table Tennis Dataset
15:30 – 15:45 Poll for Best Presentation
RA 4336
09:45 – 10:00 - Narek Moradian - Scalability and Effectiveness of Conditioning in a BDD-Based Probabilistic Database: An Empirical Evaluation
10:00 – 10:15 - Rian Sood - A Probabilistic Approach to MICE Data Imputation for Uncertain Data Management
10:15 – 10:30 - Bokyung Kim - Evidence and Scenario Modeling in Homicide Investigation: A Narrative Based Ontology
10:30 – 10:45 - Roman Ladus - When Citation Graphs Don’t Help: A Comparative Evaluation of Retrieval Interventions on Moldovan Legal Statutes
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Adam Roald Bosch - MambaLRP-IG: Explaining Mamba using Layer-Wise Relevance Propagation and Integrated Gradients
11:15 – 11:30 - Ayolt ten Have - Explainable AI for time series data
11:30 – 11:45 - Harveer Singh - Arousal Detection from Polygraphy Signals: A Channel Ablation Study on Clinical Obstructive Sleep Apnea Data
11:45 – 12:00 - Siebe Berkenbosch - Evaluation of Deep Learning Models for Blood Vessel Segmentation in CT Scan Data
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Denislav Dinev - Applying Agentic Process Mining Techniques for Synthetic Data Generation
13:30 – 13:45 - Gallyon Doest - MM-PM2: Extending the PM2 Methodology to Support Multimodal Data Sources in Process Mining
13:45 – 14:00 - Nathan Moerwanto - Process Mining for Invention Disclosure: Pipeline Development and Evaluation using Synthetic Data
14:00 – 14:15 - Yohan Kuruvilla Arikupuram - Architectural Decisions in LLM-Generated Web Applications
14:15 – 14:30 Minibreak
14:30 – 14:45 - Piotr Kondratowicz - When Does Agentic Retrieval Pay Off? A Controlled Comparison of RAG and Agent-Wrapped RAG on Multi-Hop Question Answering
14:45 – 15:00 - Elizabet Mkrtchyan - A Comparative Empirical Study of Maintainability, Reliability, and Security in AI-Generated and Human-Written Software
15:00 – 15:15 - Adham Selim - MAYA: A Collaborative Multi-Agent System for Accelerating Academic Literature Review
15:15 – 15:30 Poll for Best Presentation
Track chair: Duc le Viet
RA 3334
09:45 – 10:00 - Blert Shabani - Predicting GNSS Reliability for Autonomous Navigation Using Camera Cues
10:00 – 10:15 - Junseo Kim - Combining Recursive Weight-Sharing with Token Merging for Edge Vision Transformers
10:15 – 10:30 - Daniel Botezatu - Autonomous LLM-Driven RTL Generation for Deep Neural Networks via Multi-Agent Orchestration
10:30 – 10:45 - Art de Heer - Active Noise Control in a Laboratory Duct: Comparing Invasive and Non-Invasive Online Path Modeling
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Fabiana Zukaite - Determining Sensor Priority for Environment Identification During Indoor-Outdoor Transition
11:15 – 11:30 - J.C. Liebeton - Can Ambient WiFi Substitute for a Body-Worn IMU? Limits of Cross-Modal Substitution for Human Activity Recognition at Scale
11:30 – 11:45 - Umid Akbarov - Evaluating FPGA-Optimized Internal Self-Healing Approximate Multipliers in Vision Transformers Using TransAxx
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Benjamin Cojocaru - Neuro-Symbolic Multi-Agent System for Smart Grid Load Balancing
13:30 – 13:45 - Niek Zandt - Learning from Execution Feedback: In-Context Reinforcement Learning for Text-to-SQL on Ambiguous Real-World Databases
13:45 – 14:00 - Pedro Cecchini Said - Energy-Aware GPU Kernel Colocation
14:00 – 14:15 - Adomas Puslys - Predicting GPU Frequencies to Reduce Energy Waste
14:15 – 14:30 Minibreak
14:30 – 14:45 - Hamit Alphan Mete - Evaluation of Homomorphic Encryption Approaches in Energy Management Algorithm
14:45 – 15:00 - Aadi Malhotra - Decentralised Profile Steering for Failure-Tolerant Energy Management in Unreliable Mesh Networks
15:00 – 15:15 - Hossein Rajabali Nejad - Knowledge-Grounded Safety Risk Analysis from Textual Scenarios Using LLMs
15:15 – 15:30 - Sander Marki-Pleym - Fast Emulation of FPGA-Optimized Power-Efficient ISH- Based Approximate Multipliers in Deep Learning Pipelines
15:30 – 15:45 Poll for Best Presentation
Track chair: Tiago Prince Sales
RA 4231
09:45 – 10:00 - Loes Baart de la Faille - Architectural Patterns for Bringing Computation to Data A Systematic Mapping Study on the Data Visiting Paradigm
10:00 – 10:15 - Levi Ockels - Designing an OWL-Based Knowledge Graph Ontology for Cryptocurrency Exchange Data
10:15 – 10:30 - Mihai Grigore Santai - Internet Bad Neighbourhoods Analyser
10:30 – 10:45 - Maria Secara - Building Law Enforcement Trust in Cybercrime Data Analysis Through a Human-in-the-Loop and LLM-Supported Pipeline
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Bartosz Sanowski - Revisiting X Bot Detection in a Polish setting in the Age of AI
11:15 – 11:30 - Leonardo Iara - Encrypted Hypothesis Testing for Pearson's Correlation Coefficient under CKKS
11:30 – 11:45 - Rutger Westgeest - A Framework for Educationally-Effective Capture-The-Flag Challenges from Public-Key Cryptography Vulnerabilities
11:45 – 12:00 - Rein de Boer - Piercing the Veil of Chain-Hopping: An Automated Knowledge Graph Approach to Cross-Chain Transaction Analysis
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Marin Nofulla - Dynamic Knowledge Graph Construction via LLM-Driven Structured Extraction
13:30 – 13:45 - Ion Negru - Effect of Log-Reduction Techniques on the Performance of CAPTAIN, a Rule-Based Intrusion Detection System
13:45 – 14:00 - Daniel Jonker - A Hybrid Framework for Automatic DCAT-compliant Metadata Generation from Structured Datasets
14:00 – 14:15 - Alexandr Gidikov - Designing and Implementing a FAIR Digital Object Creation Pipeline
14:15 – 14:30 Minibreak
14:30 – 14:45 - Andrei Badea - Design and Development of an LLM-based Conversational Agent for Cyber Hygiene Support
14:45 – 15:00 - Martina Roci - What do I get from modeling?
15:00 – 15:15 - Lukáš Morvay - Designing an Ontology-Grounded LLM Assistant to Support Modeling Value Canvas Completion
15:15 – 15:30 Poll for Best Presentation
RA 4237
09:45 – 10:00 - Bram Hut - VariEdit: A Diverse Benchmarking Dataset for Knowledge Editing in LLMs
10:00 – 10:15 - Andrei-George Nadane - Measuring the Effectiveness of RAG Poisoning Attacks on Cyber Threat Intelligence Data
10:15 – 10:30 - Marlon Nelson - AI Agents for Automated Threat Modeling
10:30 – 10:45 - Gheorghe Turca - Creating a Machine-Readable SKOS Taxonomy of the NIST Cybersecurity Framework 2.0
10:45 – 11:00 Coffee Break
11:00 – 11:15 - Sundus Ali - Why Copy-Move Forgery Detectors Fail: A Transformation-Based Cross-Dataset Study Using CNN
11:15 – 11:30 - Mihnea Ambrus - Beyond the Probability Score: Evaluating Explainable AI for Document Fraud Detection in Corporate Finance
11:30 – 11:45 - Timo van den Berg - Hidden in Plain Text: Evaluating Steganographic Prompt Injection in Autonomous Coding Agents
11:45 – 12:00 - Magdalena Kaim - Prompts Carrying Secret Messages: Use of Steganography in Prompt Injection
12:00 – 13:15 Lunch Break
13:15 – 13:30 - Huy Ho - Hacking Ontology
13:30 – 13:45 - Matas Aizenas - Using Symbolic Execution to Effectively Find Panics in Rust Binaries
13:45 – 14:00 - Andrei Dorneanu - Building a FAIR Kubernetes Knowledge Graph for Cybersecurity Research
14:00 – 14:15 - Matas Tijusas - Extending GUARD: Detecting Security Smells Across IaC Languages
14:15 – 14:30 Minibreak
14:30 – 14:45 - Steef Broeder - Investigating the safety/security interactions of the Dutch transmission grid using Attack-Fault-Defense Trees
14:45 – 15:00 - Juras Macijauskas - Technological Interoperability for Machine-actionable Data Management Plans
15:00 – 15:15 - Tudor Matei - Request Cancellation in Web Application Frameworks
15:15 – 15:30 Poll for Best Presentation
Permission to make digital or hard copies of all or part of the on this website presented work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission.
Copyright University of Twente, Twente Student Conference on IT, Enschede, the Netherlands.