Project B01: A dialog-based approach to explaining machine learning models
The computer scientists and sociologists in Project B01 are investigating how dialogue-based explanations of AI systems work in real organizational contexts, such as in the medical field or in the field of predictive policing. Among other things, organizational structures, different roles, and various communication styles are taken into account. The project has already shown that dialogue-based explanations significantly improve user understanding. In the second funding phase, the team is developing a modular dialogue system that observes explanations and can adapt them specifically to the respective needs of users.
Research areas: Computer science, Sociology
Associate Member
Dr. Stefan Heindorf, Paderborn University
Former Members
Prof. Dr. Axel-Cyrille Ngonga Ngomo, Project leader
Dr. Sascha Griffiths, Research associate
Ali Manzoor, Research associate
Dimitry Mindlin, Research associate
Leonie Sieger, Research associate
Publications
Investigating Co-Constructive Behavior of Large Language Models in Explanation Dialogues
L. Fichtel, M. Spliethöver, E. Hüllermeier, P. Jimenez, N. Klowait, S. Kopp, A.-C. Ngonga Ngomo, A. Robrecht, I. Scharlau, L. Terfloth, A.-L. Vollmer, H. Wachsmuth, ArXiv:2504.18483 (2025).
Logics with probabilistic team semantics and the Boolean negation
M. Hannula, M. Hirvonen, J. Kontinen, Y. Mahmood, A. Meier, J. Virtema, Journal of Logic and Computation 35 (2025).
Facets in Argumentation: A Formal Approach to Argument Significance
J. Fichte, N. Fröhlich, M. Hecher, V. Lagerkvist, Y. Mahmood, A. Meier, J. Persson, ArXiv:2505.10982 (2025).
Investigating Co-Constructive Behavior of Large Language Models in Explanation Dialogues
L. Fichtel, M. Spliethöver, E. Hüllermeier, P. Jimenez, N. Klowait, S. Kopp, A.-C. Ngonga Ngomo, A. Robrecht, I. Scharlau, L. Terfloth, A.-L. Vollmer, H. Wachsmuth, in: Proceedings of the 26th Annual Meeting of the Special Interest Group on Discourse and Dialogue, Association for Computational Linguistics, Avignon, France, n.d.
Why not? Developing ABox Abduction beyond Repairs
A. Haak, P. Koopmann, Y. Mahmood, A.-Y. Turhan, ArXiv:2507.21955 (2025).
Dung’s Argumentation Framework: Unveiling the Expressive Power with Inconsistent Databases
Y. Mahmood, M. Hecher, A.-C. Ngonga Ngomo, in: Proceedings of the AAAI Conference on Artificial Intelligence, Association for the Advancement of Artificial Intelligence (AAAI), 2025, pp. 15058–15066.
Dung's Argumentation Framework: Unveiling the Expressive Power with Inconsistent Databases
Y. Mahmood, M. Hecher, A.-C. Ngonga Ngomo, in: 2024.
Rejection in Abstract Argumentation: Harder Than Acceptance?
J.K. Fichte, M. Hecher, Y. Mahmood, A. Meier, in: Frontiers in Artificial Intelligence and Applications, IOS Press, 2024.
Quantitative Claim-Centric Reasoning in Logic-Based Argumentation
M. Hecher, Y. Mahmood, A. Meier, J. Schmidt, in: Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence, International Joint Conferences on Artificial Intelligence Organization, 2024.
Parameterised Complexity of Consistent Query Answering via Graph Representations
T. Hankala, M. Hannula, Y. Mahmood, A. Meier, ArXiv:2412.08324 (2024).
Explaining Machines: Social Management of Incomprehensible Algorithms. Introduction
E. Esposito, Sociologica 16 (2023) 1–4.
User Involvement in Training Smart Home Agents
L.N. Sieger, J. Hermann, A. Schomäcker, S. Heindorf, C. Meske, C.-C. Hey, A. Doğangün, in: International Conference on Human-Agent Interaction, ACM, 2022.
Show all publications
