Publications

2025

1. How Large Language Models Judge and Influence Human Cooperation

   A. Pires , Laurens Samson , Sennay Ghebreab, and 1 more author

   arXiv preprint, 2025

   Under Review

   Abs

   We investigate how Large Language Models evaluate and influence patterns of human cooperation, with implications for AI-human interaction design.

2. Privacy Aware Visual Language Models

   Laurens Samson , N. Barazani , Sennay Ghebreab, and 1 more author

   arXiv preprint, 2025

   Under Review

   Abs

   We present a novel approach to visual language models that preserves privacy while maintaining performance, crucial for deployment in public and sensitive settings.

2023

1. When ’doing ethics’ meets public procurement of smart city technology – an Amsterdam case study

   M. De Kreek , K. Alfrink , M. Waal, and 4 more authors

   In IASDR 2023: Life-Changing Design, 2023

   Abs

   We examine the intersection of ethical AI frameworks and public procurement processes in smart city initiatives, using Amsterdam as a case study.

2022

1. Generative Adversarial Networks for Image Segmentation

   M. Ghafoorian , Laurens Samson , and O. Booij

   2022

   Abs

   Novel adversarial training method for improving image segmentation accuracy and robustness in autonomous driving applications.

2021

1. Back to Basics: Deep Reinforcement Learning in Traffic Signal Control

   Sierk Kanis , Laurens Samson , D. Bloemberen, and 1 more author

   In 29th International Conference on Advances in Geographic Information Systems (SIGSPATIAL ’21), 2021

   Abs

   We demonstrate that simple deep reinforcement learning approaches can outperform complex methods in traffic signal control when properly implemented.

2019

1. I Bet You Are Wrong: Gambling Adversarial Networks for Structured Semantic Segmentation

   Laurens Samson , N. Noord , O. Booij, and 3 more authors

   In IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2019

   Abs PDF

   We propose a novel adversarial training framework for structured semantic segmentation that improves robustness and accuracy in autonomous driving scenarios.