74 citations · 140 across the 34 of their papers we have counts for
38 papers
What Pixels Are Enough? SEAMS: Sufficiency Saliency via MSE-Preservation Soft-Masks
Magdalena Trędowicz, Łukasz Struski, Arkadiusz Lewicki +4
Saliency maps are most useful when they identify the image regions that are sufficient to preserve a model's behaviour. We introduce SEAMS, a sufficiency-based saliency method that…
SoftMoE: Soft Differentiable Routing for Mixture-of-Experts in LLMs
Mikołaj Zasada, Łukasz Struski, Jacek Tabor +1
Sparse Mixture-of-Experts (MoE) architectures enable scaling LLM parameters under a fixed inference budget by activating only a small subset of experts via top- routing. While t…
LAPLEX: The FFT of Learnable Laplace Kernels
Łukasz Struski, Hanna Blazhko, Piotr Kubaty +1
Fast linear algebra in deep learning usually comes with a choice: fixed geometry and exact computation, as in the Fourier transform, or adaptive geometry paid for by dense paramete…
Conceptualizing Embeddings: Sparse Disentanglement for Vision-Language Models
Piotr Kubaty, Patryk Marszałek, Łukasz Struski +3
Vision-language models learn powerful multimodal embeddings, yet their internal semantics remain opaque. While sparse autoencoders (SAEs) can extract interpretable features, they r…
ProDG: Prototypes for Data-Free Generative Post-Hoc Explainability
Piotr Borycki, Magdalena Trędowicz, Jacek Tabor +2
Ante-hoc interpretability methods based on prototypes provide highly accurate explanations by utilizing the intuitive "this looks like that" reasoning paradigm. On the other hand,…
Stop Marginalizing My Dreams: Model Inversion via Laplace Kernel for Continual Learning
Patryk Krukowski, Jacek Tabor, Przemysław Spurek +2
Data-free continual learning (DFCIL) relies on model inversion to synthesize pseudo-samples and mitigate catastrophic forgetting. However, existing inversion methods are fundamenta…
SIDE: Sparse Information Disentanglement for Explainable Artificial Intelligence
Viktar Dubovik, Łukasz Struski, Jacek Tabor +1
Understanding the decisions made by deep neural networks is essential in high-stakes domains such as medical imaging and autonomous driving. Yet, these models often lack transparen…
EPIC: Explanation of Pretrained Image Classification Networks via Prototype
Piotr Borycki, Magdalena Trędowicz, Szymon Janusz +4
Explainable AI (XAI) methods generally fall into two categories. Post-hoc approaches generate explanations for pre-trained models and are compatible with various neural network arc…
LapSum -- One Method to Differentiate Them All: Ranking, Sorting and Top-k Selection
Łukasz Struski, Michał B. Bednarczyk, Igor T. Podolak +1
We present a novel technique for constructing differentiable order-type operations, including soft ranking, soft top-k selection, and soft permutations. Our approach leverages an e…
SEMU: Singular Value Decomposition for Efficient Machine Unlearning
Marcin Sendera, Łukasz Struski, Kamil Książek +3
While the capabilities of generative foundational models have advanced rapidly in recent years, methods to prevent harmful and unsafe behaviors remain underdeveloped. Among the pre…
InfoDisent: Explainability of Image Classification Models by Information Disentanglement
Łukasz Struski, Dawid Rymarczyk, Jacek Tabor
In this work, we introduce InfoDisent, a hybrid approach to explainability based on the information bottleneck principle. InfoDisent enables the disentanglement of information in t…
PrAViC: Probabilistic Adaptation Framework for Real-Time Video Classification
Magdalena Trędowicz, Marcin Mazur, Szymon Janusz +3
Video processing is generally divided into two main categories: processing of the entire video, which typically yields optimal classification outcomes, and real-time processing, wh…
ProPML: Probability Partial Multi-label Learning
Łukasz Struski, Adam Pardyl, Jacek Tabor +1
Partial Multi-label Learning (PML) is a type of weakly supervised learning where each training instance corresponds to a set of candidate labels, among which only some are true. In…
Interpretability Benchmark for Evaluating Spatial Misalignment of Prototypical Parts Explanations
Mikołaj Sacha, Bartosz Jura, Dawid Rymarczyk +3
Prototypical parts-based networks are becoming increasingly popular due to their faithful self-explanations. However, their similarity maps are calculated in the penultimate networ…
ProMIL: Probabilistic Multiple Instance Learning for Medical Imaging
Łukasz Struski, Dawid Rymarczyk, Arkadiusz Lewicki +3
Multiple Instance Learning (MIL) is a weakly-supervised problem in which one label is assigned to the whole bag of instances. An important class of MIL models is instance-based, wh…
r-softmax: Generalized Softmax with Controllable Sparsity Rate
Klaudia Bałazy, Łukasz Struski, Marek Śmieja +1
Nowadays artificial neural network models achieve remarkable results in many disciplines. Functions mapping the representation provided by the model to the probability distribution…
ProPaLL: Probabilistic Partial Label Learning
Łukasz Struski, Jacek Tabor, Bartosz Zieliński
Partial label learning is a type of weakly supervised learning, where each training instance corresponds to a set of candidate labels, among which only one is true. In this paper,…
Bounding Evidence and Estimating Log-Likelihood in VAE
Łukasz Struski, Marcin Mazur, Paweł Batorski +2
Many crucial problems in deep learning and statistical inference are caused by a variational gap, i.e., a difference between model evidence (log-likelihood) and evidence lower boun…
Interpretable Image Classification with Differentiable Prototypes Assignment
Dawid Rymarczyk, Łukasz Struski, Michał Górszczak +3
We introduce ProtoPool, an interpretable image classification model with a pool of prototypes shared by the classes. The training is more straightforward than in the existing metho…
MisConv: Convolutional Neural Networks for Missing Data
Marcin Przewięźlikowski, Marek Śmieja, Łukasz Struski +1
Processing of missing data by modern neural networks, such as CNNs, remains a fundamental, yet unsolved challenge, which naturally arises in many practical applications, like image…
Efficient GPU implementation of randomized SVD and its applications
Łukasz Struski, Paweł Morkisz, Przemysław Spurek +2
Matrix decompositions are ubiquitous in machine learning, including applications in dimensionality reduction, data compression and deep learning algorithms. Typical solutions for m…
Pharmacoprint -- a combination of pharmacophore fingerprint and artificial intelligence as a tool for computer-aided drug design
Dawid Warszycki, Łukasz Struski, Marek Śmieja +2
Structural fingerprints and pharmacophore modeling are methodologies that have been used for at least two decades in various fields of cheminformatics: from similarity searching to…
Flow-based SVDD for anomaly detection
Marcin Sendera, Marek Śmieja, Łukasz Maziarka +3
We propose FlowSVDD -- a flow-based one-class classifier for anomaly/outliers detection that realizes a well-known SVDD principle using deep learning tools. Contrary to other appro…
HyperPocket: Generative Point Cloud Completion
Przemysław Spurek, Artur Kasymov, Marcin Mazur +5
Scanning real-life scenes with modern registration devices typically give incomplete point cloud representations, mostly due to the limitations of the scanning process and 3D occlu…