Faculty of Mathematics, Physics
and Informatics
Comenius University Bratislava

Plenary Presentations

Student Section (10:15 - 11:15)

Enmily Loelbis Garcia Angulo: Optical Emission Spectroscopy of Isoflurane by Electron Impact

Co-authors: 

Time of proceedings: 10:15 hod. 

Abstract: 

Aneta Kolodzejová: Development and Initial Testing of a Physics-Specific Critical Thinking Assessment Tool

Co-authors: 

Time of proceedings: 10:30 hod. 

Abstract: 

Neda Bagheri Renani: Numerical Solution of the Two-Dimensional Bates Model Compact Finite Differences and Finite Element Comparison

Co-authors: 

Time of proceedings: 10:45 hod. 

Abstract: 

Ján Pastorek: Mathematical Enterprise: Automated Exploration and Discovery at Scale

Co-authors: 

Time of proceedings: 11:00 hod. 

Abstract: 

Science results at FMFI (13:15 - 15:15)

 

Rastislav Kráľovič: Agents on a line: implementing memory by finite-state agents

Time of proceedings: 13:15

Abstract: 

Peter Maták, Tomáš Blažek, Ján Ramaj, Martina Sabová: 
Towards the minimal effective theory for leptogenesis, dark matter, and neutrino masses

Time of proceedings: 13:45

Abstract: We present a bottom-up approach to an effective theory that simultaneously explains the matter–antimatter asymmetry via leptogenesis, the dark matter relic abundance via freeze-in or freeze-out mechanisms, and neutrino masses via the Weinberg operator. We show that, in the minimal scenario, only two new particles and a single portal operator coupling the visible and dark sectors are sufficient beyond the Standard Model. This contribution is primarily based on arXiv:2504.15164.

Peter Vankúš, Michaela Vargová:
Innovative Approaches to Teaching Probability in Secondary Schools

Time of proceedings: 14:15

Abstract: In this paper, we present innovative approaches to teaching probability that were developed as part of the KEGA 037UK-4/2024 project Innovative learning technologies in the preparation of future mathematics teachers and the project Digital Transformation of Education and Schools (DiTEdu, code ITMS2014+: 401402DVR6). These approaches aim to foster a deep conceptual understanding of probabilistic concepts through their gradual constructivist development, with an emphasis on visualization, manipulative activities, and experimentation. The topics covered include conditional probability, geometric probability, and expected value.

Vladimír Boža, Vladimír Macko:
Addition is almost all you need: Compressing neural networks with double binary factorization

Time of proceedings: 14:45

Abstract: Binary quantization approaches, which replace weight matrices with binary matrices and substitute costly multiplications with cheaper additions, offer a computationally efficient approach to address the increasing computational and storage requirements of Large Language Models (LLMs). However, the severe quantization constraint (±1) can lead to significant accuracy degradation. In this paper, we propose Double Binary Factorization (DBF), a novel method that factorizes dense weight matrices into products of two binary (sign) matrices, each accompanied by scaling vectors. DBF preserves the efficiency advantages of binary representations while achieving compression rates that are competitive with or superior to state-of-the-art methods. Specifically, in a 1-bit per weight range, DBF is better than existing binarization approaches. In a 2-bit per weight range, DBF is competitive with the best quantization methods like QuIP# and QTIP. Unlike most existing compression techniques, which offer limited compression level choices, DBF allows fine-grained control over compression ratios by adjusting the factorization's intermediate dimension. Based on this advantage, we further introduce an algorithm for estimating non-uniform layer-wise compression ratios for DBF, based on previously developed channel pruning criteria.