activity
20152026
most citedNon-Gaussian Gaussian Processes for Few-Shot Regression

5 citations · 26 across the 53 of their papers we have counts for

collaborators
Showing cs.LGShow all

30 papers · 1 filter

cs.LG2026

GROM: Gradient-Free Rapid One-Shot Machine Unlearning

Paweł Batorski, Przemysław Spurek, Paul Swoboda

Machine unlearning has become a critical capability for safely removing specific, sensitive knowledge from large language models (LLMs). Current state-of-the-art approaches primari…

cs.LG2026

Spectral Energy Centroid: a Metric for Improving Performance and Analyzing Spectral Bias in Implicit Neural Representations

Tomasz Dądela, Adam Kania, Maciej Rut +1

Implicit Neural Representations (INRs) model continuous signals using multilayer perceptrons (MLPs), enabling compact, differentiable, and high-fidelity representations of data acr…

cs.LG2026

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…

cs.LG2026

SoftSAE: Dynamic Top-K Selection for Adaptive Sparse Autoencoders

Jakub Stępień, Marcin Mazur, Jacek Tabor +1

Sparse Autoencoders (SAEs) have become an important tool in mechanistic interpretability, helping to analyze internal representations in both Large Language Models (LLMs) and Visio…

cs.LG2026

REBEL: Hidden Knowledge Recovery via Evolutionary-Based Evaluation Loop

Patryk Rybak, Paweł Batorski, Paul Swoboda +1

Machine unlearning for LLMs aims to remove sensitive or copyrighted data from trained models. However, the true efficacy of current unlearning methods remains uncertain. Standard e…

cs.LG2025

Memory Self-Regeneration: Uncovering Hidden Knowledge in Unlearned Models

Agnieszka Polowczyk, Alicja Polowczyk, Joanna Waczyńska +2

The impressive capability of modern text-to-image models to generate realistic visuals has come with a serious drawback: they can be misused to create harmful, deceptive or unlawfu…