Tomasz Kuśmierczyk: About me

I am a machine learning researcher working on probabilistic modeling, Bayesian deep learning, and reliable AI systems. I am interested in bridging statistical inference and modern deep learning to develop models that are more robust, interpretable, and effective in real-world settings.

Contact

e-mail: tomasz [dot] kusmierczyk [at] gmail.com

My Curriculum Vitae

Media

GitHub
LinkedIn
Scholar
DBLP

2024-2026 Project
ResearchGate
News (Twitter)

Publications

V. Dubovik, P. Marszałek, J. Tabor, T. Kuśmierczyk: Bayesian Fine-tuning in Projected Subspaces. arXiv github
P. Marszałek, T. Kuśmierczyk, M. Śmieja: TACTIC for Navigating the Unknown: Tabular Anomaly deteCTion via In-Context inference. arXiv github
Lakshmana Sri Harsha Nemani, PK Srijith, T. Kuśmierczyk: Efficient Uncertainty in LLMs through Evidential Knowledge Distillation. arXiv github
P. Marszałek, T. Kuśmierczyk*, W. Wydmański, J. Tabor, M. Śmieja*: ZEUS: Zero-shot Embeddings for Unsupervised Separation of Tabular Data. NeurIPS 2025 (CORE rank: A*) arXiv proceedings github
P. Marszałek*, K. Bałazy, J. Tabor, T. Kuśmierczyk*: Minimal Ranks, Maximum Confidence: Parameter-efficient Uncertainty Quantification for LoRA. EMNLP 2025 (CORE rank: A*) arXiv proceedings github
M. Sendera*, A. Sorkhei, T. Kuśmierczyk*: Revisiting the Equivalence of Bayesian Neural Networks and Gaussian Processes: On the Importance of Learning Activations. UAI 2025 (CORE rank: A) arXiv proceedings github NeurIPS 2024 Workshop
P. Borycki, P. Kubacki, M. Przewięźlikowski, T. Kuśmierczyk, J. Tabor, P. Spurek: Hypernetwork Approach to Bayesian MAML (Student Abstract). AAAI 2025. proceedings arXiv
E. de Souza da Silva, T. Kuśmierczyk, M. Hartmann, A. Klami: Prior Specification for Bayesian Matrix Factorization via Prior Predictive Matching. JMLR 2023. arXiv CODE ONLINE proceedings
T. Kuśmierczyk, A. Klami: Reliable Categorical Variational Inferencewith Mixture of Discrete Normalizing Flows. arXiv CODE SLIDES
O. Nyberg, T. Kuśmierczyk, A. Klami: Uplift Modeling with High Class Imbalance. ACML 2021. VIDEO proceedings
T. Kuśmierczyk, J. Sakaya, A. Klami: Correcting Predictions for Approximate Bayesian Inference. AAAI 2020. (CORE rank: A*) arXiv CODE POSTER proceedings
T. Kuśmierczyk, J. Sakaya, A. Klami: Variational Bayesian Decision-making for Continuous Utilities. NeurIPS 2019. (CORE rank: A*) CODE arXiv SLIDES POSTER proceedings
C. Trattner, T. Kuśmierczyk, K. Nørvåg: Investigating and Predicting Online Food Recipe Upload Behavior. Information Processing & Management. (IF: 3.4) PDF ONLINE
T. Kuśmierczyk, M. Gomez-Rodriguez: On the Causal Effect of Badges. WWW 2018. (CORE rank: A*) CODE arXiv SLIDES
T. Kuśmierczyk, K. Nørvåg: On Validation and Predictability of Digital Badges' Influence on Individual Users. AAAI 2018. (CORE rank: A*) CODE arXiv POSTER
T. Kuśmierczyk, K. Nørvåg: Online Food Recipe Title Semantics: Combining Nutrient Facts and Topics. CIKM 2016. (CORE rank: A) CODE PDF POSTER
T. Kuśmierczyk, C. Trattner, K. Nørvåg: Understanding and Predicting Online Food Recipe Production Patterns. Hypertext 2016. (CORE rank: A) PDF SLIDES ONLINE
G. Bokota, M. Magnowska, T. Kuśmierczyk, M. Lukasik, et al.: Computational approach to dendritic spine taxonomy and shape transition analysis. Frontiers in Computational Neuroscience, 2016. (IF:2.3) ONLINE CODE
M. Rokicki, E. Herder, T. Kuśmierczyk, C. Trattner: Plate and Prejudice: Gender Differences in Online Cooking. UMAP 2016. (CORE rank: B) PDF
T. Kuśmierczyk, K. Nørvåg: Mining Correlations on Massive Bursty Time Series Collections. DASFAA 2015. (CORE rank: B) CODE PDF SLIDES
T. Kuśmierczyk, C. Trattner, K. Nørvåg: Temporality in Online Food Recipe Consumption and Production. WWW 2015. (CORE rank: A*) PDF POSTER
T. Kuśmierczyk, C. Trattner, K. Nørvåg: Temporal Patterns in Online Food Innovation. TempWeb 2015 (WWW Proceedings). PDF SLIDES
W. Kosinski, M. Sydow, T. Kuśmierczyk, P. Rembelski: Application of Ant-Colony Optimisation to Compute Diversified Entity Summarisation on Semantic Knowledge Graphs. FedCSIS 2013: 69-76. PDF
T. Kuśmierczyk, M. Sydow: Towards a Keyword-Focused Web Crawler. IIS 2013: 187-197. PDF
T. Kuśmierczyk, M. Lukasik, L. Bolikowski, H. Son Nguyen: Comparing Hierarchical Mathematical Document Clustering against the Mathematics Subject Classification Tree. Intelligent Tools for Building a Scientific Information Platform 2013: 365-392. PDF
M. Lukasik, T. Kuśmierczyk, L. Bolikowski, H. Son Nguyen: Hierarchical, Multi-label Classification of Scholarly Publications: Modifications of ML-KNN Algorithm. Intelligent Tools for Building a Scientific Information Platform 2013: 343-363. PDF

Tomasz Kuśmierczyk: About me

Contact

My Curriculum Vitae

Media

Publications

Code contributions

Presentations

Parameter-efficient Uncertainty Quantification for LoRA (EMNLP 2025)

Beyond BBB: Practical Alternatives to Posterior Approximation in Bayesian Neural Networks

Introduction to Bayesian methods and Bayesian Neural Networks

Priors for BNNs

Overconfidence and subnetwork Inference for BNNs

Introduction to modern Variational Inference

Mixture of Discrete Normalizing Flows for Variational Inference

Loss-calibrated Variational Inference

Variational inference using implicit distributions

On the Causal Effect of Digital Badges (WWW2018)

What are the negative effects of social media?: fighting fake information (Defense Trial Lecture)

Sampling and Markov Chain Monte Carlo Techniques (Self-study course)

Probabilistic Models in Recommender Systems: Time Variant Models (Self-study course)

Temporal Patterns in Online Food Innovation (TempWeb2015)

Mining Correlations on Massive Bursty Time Series Collection (DASFAA2015)