Kristian Kersting

Short info

My Artificial Intelligence and Machine Learning lab would like to make computers learn so much about the world, so rapidly and flexibly, as humans. This poses many deep and fascinating scientific problems: How can computers learn with less help from us and data? How can computers reason about and learn with complex data such as graphs and uncertain databases? How can pre-existing knowledge be exploited? How can computers decide autonomously which representation is best for the data at hand? Can learned results be physically plausible or be made understandable by us? How can computers learn together with us in the loop? To this end, we develop novel machine learning (ML) and artificial intelligence (AI) methods, i.e., novel computational methods that contain and combine for example search, logical and probabilistic techniques as well as (deep) (un)supervised and reinforcement learning methods.

Open Science

Brack, M., Katakol, S., Friedrich, F., Schramowski, P., Ravi, H., Kersting, K., & Kale, A. (2025).
How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions.
arXiv preprint, 2506.16679.

Busch, F. P., Kamath, R., Mitchell, R., Stammer, W., Kersting, K., & Mundt, M. (2024).
Where is the truth? the risk of getting confounded in a continual world.
arXiv preprint arXiv: 2402.06434.

Deiseroth, B., Brack, M., Schramowski, P., Kersting, K., & Weinbach, S. (2024).
T-FREE: Subword tokenizer-free generative LLMs via sparse representations for memory-efficient embeddings.
arXiv preprint arXiv: 2406.19223.

Delfosse, Q., Blüml, J., Tatai, F., Vincent, T., Gregori, B., Dillies, E., ... & Kersting, K. (2025).
Deep Reinforcement Learning Agents are not even close to Human Intelligence.
arXiv preprint arXiv:2505.21731.

Friedrich, F., Tedeschi, S., Schramowski, P., Brack, M., Navigli, R., Nguyen, H., ... & Kersting, K. (2025).
Llms lost in translation: M-alert uncovers cross-linguistic safety gaps.
ICLR 2025 Workshop on Building Trust in Language Models and Applications.

Friedrich, F., Welsch, T. G., Brack, M., Schramowski, P., & Kersting, K. (2025).
Beyond Overcorrection: Evaluating Diversity in T2I Models with DIVBENCH.
arXiv preprint: 2507.03015.

Härle, R., Friedrich, F., Brack, M., Deiseroth, B., Schramowski, P., & Kersting, K. (2024).
SCAR: Sparse conditioned autoencoders for concept detection and steering in LLMs.
arXiv preprint arXiv: 2411.07122.

Härle, R., Friedrich, F., Brack, M., Wäldchen, S., Deiseroth, B., Schramowski, P., & Kersting, K. (2025).
Measuring and Guiding Monosemanticity.
arXiv preprint, 2506.19382.

Helff, L., Friedrich, F., Brack, M., Kersting, K., & Schramowski, P. (2024).
Llavaguard: An open vlm-based framework for safeguarding vision datasets and models.
arXiv preprint arXiv: 2406.05113.

Kamath, R., Mitchell, R., Paul, S., Kersting, K., & Mundt, M. (2024).
BOWL: A Deceptively Simple Open World Learner.
arXiv preprint arXiv: 2402.04814.

Kraus, M., Divo, F., Steinmann, D., Dhami, D. S., & Kersting, K. (2024) (2024).
United We Pretrain, Divided We Fail! Representation Learning for Time Series by Pretraining on 75 Datasets at Once.
arXiv preprint arXiv: 2402.15404.

Kraus, M., Steinmann, D., Wüst, A., Kokozinski, A., & Kersting, K. (2024).
Right on time: Revising time series models by constraining their explanations.
arXiv preprint:2402.12921

Poonia, H., Willig, M., Yu, Z., Zečević, M., Kersting, K., & Dhami, D. S. (2024).
$\chi $ SPN: Characteristic Interventional Sum-Product Networks for Causal Inference in Hybrid Domains.
arXiv preprint arXiv: 2408.07545.

Schuhmann, C., Kaczmarczyk, R., Rabby, G., Friedrich, F., Kraus, M., Kalyan, K., Kersting, K., & Auer, S. (2025).
EmoNet-Face: An expert-annotated benchmark for synthetic emotion recognition.
arXiv preprint arXiv: 2505.20033.

Stammer, W., Friedrich, F., Steinmann, D., Brack, M., Shindo, H., & Kersting, K. (2024).
Learning by self-explaining.
arXiv preprint arXiv:2309.08395.

Steinmann, D., Divo, F., Kraus, M., Wüst, A., Struppek, L., Friedrich, F., & Kersting, K. (2024).
Navigating shortcuts, spurious correlations, and confounders: From origins via detection to mitigation.
arXiv preprint arXiv:2412.05152.

Steinmann, D., Stammer, W., Friedrich, F., & Kersting, K. (2024).
Learning to intervene on concept bottlenecks.
arXiv preprint arXiv:2308.13453.

Steinmann, D., Stammer, W., Wüst, A., & Kersting, K. (2025).
Object Centric Concept Bottlenecks.
arXiv preprint: 2505.24492.

Woydt, T., Willig, M., Wüst, A., Helff, L., Stammer, W., Rothkopf, C. A., & Kersting, K. (2025).
Fodor and Pylyshyn's Legacy-Still No Human-like Systematic Compositionality in Neural Networks.
arXiv preprint arXiv:2506.01820

Wüst, A., Woydt, T., Helff, L., Ibs, I., Stammer, W., Dhami, D. S., ... & Kersting, K. (2024).
Bongard in Wonderland: Visual Puzzles that Still Make AI Go Mad?.
arXiv preprint arXiv: 2410.19546.

Ye, Z., Arenz, O., & Kersting, K. (2025).
Learning from Less: Guiding Deep Reinforcement Learning with Differentiable Symbolic Planning.
arXiv preprint arXiv: 2505.11661.

Yu, Z., Dhami, D. S., & Kersting, K. (2021).
Sum-Product-Attention Networks: Leveraging Self-Attention in Probabilistic Circuits.
arXiv preprint

Zečević, M., Dhami, D. S., Rothkopf, C. A., & Kersting, K. (2021).
Structural Causal Interpretation Theorem.
arXiv preprint

Articles

Bellagente, M., Brack, M., Teufel, H., Friedrich, F., Deiseroth, B., Eichenberg, C., Kersting, K., ... & Weinbach, S. (2024).
Multifusion: Fusing pre-trained models for multi-lingual, multi-modal image generation.
Advances in Neural Information Processing Systems, 36.

Brack, M., Friedrich, F., Hintersdorf, D., Struppek, L., Schramowski, P., & Kersting, K. (2024).
SEGA: Instructing text-to-image models using semantic guidance.
Advances in Neural Information Processing Systems, 36.

Brack, M., Schramowski, P., Deiseroth, B., & Kersting, K. (2023).
Illume: Rationalizing vision-language models through human interactions.

Carter, A. A., & Kaiser, D. (2024).
An object numbering task reveals an underestimation of complexity for typically structured scenes.
Psychonomic Bulletin & Review, 1-10.

Deiseroth, B., Deb, M., Weinbach, S., Brack, M., Schramowski, P., & Kersting, K. (2024).
Atman: Understanding transformer predictions through memory efficient attention manipulation.
Advances in Neural Information Processing Systems, 36.

Friedrich, F., Hämmerl, K., Schramowski, P., Brack, M., Libovický, J., Fraser, A., & Kersting, K. (2025).
Multilingual text-to-image generation magnifies gender stereotypes.
Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics, 19656-19679.

Friedrich, F., Stammer, W., Schramowski, P., & Kersting, K. (2023).
A typology for exploring the mitigation of shortcut behaviour.
Nature Machine Intelligence, 5(3), 319-330.

Friedrich, F., Stammer, W., Schramowski, P., Kersting, K. (2023).
Revision Transformers: Instructing Language Models to Change their Values. In
Proceedings of the 26th European Conference on Artificial Intelligence (ECAI).

Pabst, K., Gkanias, E., Webb, B., Homberg, U., & Endres, D. (2024).
A computational model for angular velocity integration in a locust heading circuit.
PLOS Computational Biology, 20(12), e1012155.

Pfanschilling, V., Shindo, H., Dhami, D. S., & Kersting, K. (2025) (2025).
NeST: The neuro-symbolic transpiler.
International Journal of Approximate Reasoning, 179, 109369.

Schramowski, P., Brack, M., Deiseroth, B., & Kersting, K. (2023).
Safe latent diffusion: Mitigating inappropriate degeneration in diffusion models. In
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(pp. 22522-22531).

Schramowski, P., Turan, C., Andersen, N., Rothkopf, C. A., & Kersting, K. (2022).
Large pre-trained language models contain human-like biases of what is right and wrong to do.
Nature Machine Intelligence, 4(3), 258-268.

Shindo, H., Brack, M., Sudhakaran, G., Dhami, D. S., Schramowski, P., & Kersting, K. (2024).
Deisam: Segment anything with deictic prompting.
Advances in Neural Information Processing Systems, 37, 52266-52295.

Shindo, H., Pfanschilling, V., Dhami, D. S., & Kersting, K. (2023).
α ILP: thinking visual scenes as differentiable logic programs.
Machine Learning, 112(5), 1465-1497.

Shindo, H., Pfanschilling, V., Dhami, D. S., & Kersting, K. (2024).
Learning differentiable logic programs for abstract visual reasoning.
Machine Learning, 113(11), 8533-8584.

Skryagin, A., Ochs, D., Dhami, D. S., & Kersting, K. (2023).
Scalable Neural-Probabilistic Answer Set Programming.
Journal of Artificial Intelligence Research (JAIR) 78:579–617.

Stammer, W., Wüst, A., Steinmann, D., & Kersting, K. (2024).
Neural concept binder.
Advances in Neural Information Processing Systems (37), 71792-71830.

Thoma, N., Yu, Z., Ventola, F., & Kersting, K. (2021).
RECOWNs: Probabilistic Circuits for Trustworthy Time Series Forecasting.
Proceedings of the 4th Workshop on Tractable Probabilistic Modeling (TPM 2021).

Thomas, T., Straub, D., Tatai, F., Shene, M., Tosik, T., Kersting, K., & Rothkopf, C. A. (2024).
Modelling dataset bias in machine-learned theories of economic decision-making.
Nature Human Behaviour, 1-13.

Ventola, F., Dhami, D. S., & Kersting, K. (2021).
Generative Clausal Networks: Relational Decision Trees as Probabilistic Circuits.
Proceedings of the 30th International Conference on Inductive Logic Programming (ILP).

Yu, Z., Trapp, M., & Kersting, K. (2024).
Characteristic Circuits.
Advances in Neural Information Processing Systems, 36.

Yu, Z., Ventola, F. G., & Kersting, K. (2021).
Whittle networks: A deep likelihood model for time series.
International Conference on Machine Learning (PMLR), 12177-12186.

Yu, Z., Zhu, M., Trapp, M., Skryagin, A., & Kersting, K. (2021).
Leveraging Probabilistic Circuits for Nonparametric Multi-Output Regression.
Proceedings of the 37th Conference on Uncertainty in Artificial Intelligence (UAI 2021).