Institution profile

Slovak University of Technology

Academic institutioneurope · sk
Official website
Research library18linked papers
Opportunities0open roles
Selected work

Representative Papers

Deep Dictionary-Free Method for Identifying Linear Model of Nonlinear System with Input Delay

Jun 03, 2025International Conference on Process Control

Modeling nonlinear systems with input delays remains challenging due to the failure of conventional linear methods and the difficulty of constructing appropriate basis function dictionaries. To address this, we propose an LSTM-enhanced, dictionary-free deep Koopman framework. Our method eliminates reliance on predefined basis dictionaries by leveraging LSTM networks to automatically learn latent temporal dependencies between past inputs and states, thereby enabling linear approximation of nonlinear dynamics in a learned embedding space. Crucially, input delays are implicitly encoded within the Koopman operator learning process, obviating explicit delay feature engineering. Experimental results demonstrate that our approach achieves significantly higher prediction accuracy than extended dynamic mode decomposition (eDMD) on unknown nonlinear systems, while matching eDMD’s performance on systems with known dynamics—indicating strong generalization capability and robustness.

1 citationsRead paper

MAECO-Lite: Modular Ontology for Dynamic Malware Analysis

May 29, 2026

This study addresses the semantic ambiguity and reasoning challenges in existing dynamic malware analysis standards—such as MAEC and STIX—stemming from their conflation of persistent artifacts with runtime events. To resolve this, the work introduces the Unified Foundational Ontology (UFO) into the domain for the first time and proposes MAECO-Lite, a lightweight, modular ontology that clearly delineates core concepts including malware samples, processes, actions, system artifacts, and MITRE ATT&CK techniques. Crucially, MAECO-Lite enforces a strict ontological separation between persistent entities and runtime events. Empirical validation using description logic-based concept learning algorithms demonstrates that MAECO-Lite significantly enhances reasoning performance while preserving semantic rigor, thereby achieving an effective balance between computational tractability and ontological clarity.

0 citationsRead paper

What Would GPT Click: Practical Effects of Human-AI Behavioral Misalignment and the Cost of Synthetic Participants in User Experience

May 18, 2026

This study addresses the systematic discrepancies between synthetic user behaviors generated by large language models (LLMs) and real human actions, which threaten the validity of user experience (UX) research. Conducting first-click tests across 12 authentic UX tasks with 3,431 participants, this work presents the first quantitative evaluation of GPT’s ability to predict both human click behavior and underlying reasoning processes. Synthetic responses were generated using role prompting, chain-of-thought instructions, and varied sampling parameters, then compared against empirical click data. Results reveal that in 53% of tasks, GPT’s predicted distributions significantly diverge from actual human behavior. The findings indicate that cognitive biases stemming from LLMs’ inherent statistical properties are not easily mitigated through prompt engineering; current optimization strategies enhance surface-level plausibility without improving behavioral fidelity, thereby highlighting critical limitations in deploying such models for UX decision-making.

0 citationsRead paper
Recent publications

Latest Papers

MAECO-Lite: Modular Ontology for Dynamic Malware Analysis

May 29, 2026

This study addresses the semantic ambiguity and reasoning challenges in existing dynamic malware analysis standards—such as MAEC and STIX—stemming from their conflation of persistent artifacts with runtime events. To resolve this, the work introduces the Unified Foundational Ontology (UFO) into the domain for the first time and proposes MAECO-Lite, a lightweight, modular ontology that clearly delineates core concepts including malware samples, processes, actions, system artifacts, and MITRE ATT&CK techniques. Crucially, MAECO-Lite enforces a strict ontological separation between persistent entities and runtime events. Empirical validation using description logic-based concept learning algorithms demonstrates that MAECO-Lite significantly enhances reasoning performance while preserving semantic rigor, thereby achieving an effective balance between computational tractability and ontological clarity.

0 citationsRead paper

What Would GPT Click: Practical Effects of Human-AI Behavioral Misalignment and the Cost of Synthetic Participants in User Experience

May 18, 2026

This study addresses the systematic discrepancies between synthetic user behaviors generated by large language models (LLMs) and real human actions, which threaten the validity of user experience (UX) research. Conducting first-click tests across 12 authentic UX tasks with 3,431 participants, this work presents the first quantitative evaluation of GPT’s ability to predict both human click behavior and underlying reasoning processes. Synthetic responses were generated using role prompting, chain-of-thought instructions, and varied sampling parameters, then compared against empirical click data. Results reveal that in 53% of tasks, GPT’s predicted distributions significantly diverge from actual human behavior. The findings indicate that cognitive biases stemming from LLMs’ inherent statistical properties are not easily mitigated through prompt engineering; current optimization strategies enhance surface-level plausibility without improving behavioral fidelity, thereby highlighting critical limitations in deploying such models for UX decision-making.

0 citationsRead paper

Distorted Perspectives of LLM-Simulated Preferences: Can AI Mislead Design?

May 18, 2026

This study investigates the alignment between large language models (LLMs) simulating user visual design preferences and actual human preferences, revealing systematic biases and risks of misguidance in AI-assisted design. Drawing on 29 real-world user tests from the UXtweak platform (n=2073), the authors conduct multimodal simulation experiments by manipulating LLM reasoning strategies, sampling approaches, role assignments, and prompt specificity. For the first time in authentic design contexts, they systematically demonstrate that LLM-generated simulations exhibit significant and consistent deviations from real user judgments, producing explanations that lack genuine perceptual grounding and semantic depth while over-relying on superficial, generalized features. This work establishes an empirical foundation and methodological framework for evaluating the algorithmic fidelity of LLMs in human–computer interaction scenarios.

0 citationsRead paper