arXiv subjects

Maria Ganzha

Publications and source records attributed to Maria Ganzha.

At least 19 recordsLinked to original sources

arXiv2026

Semantic Radiance Fields as Simulators for Spatial Reasoning in Real-World Scenes

Training and evaluating spatial reasoning in embodied agents requires diverse environments that are both geometrically faithful and semantically queryable. Synthetic simulators offer ground truth semantics but sacrifice realism; simulators based on reconstructions of real-world environments have realistic appearance but lack ground truth semantics by default. We propose using Semantic Radiance Fields (SRF) as simulators for spatial reasoning agents. SRFs are a representation that unifies these requirements by lifting 2D semantic segmentations from pretrained vision models into a 3D radiance field that jointly encodes geometry, appearance, and per-class semantic identity. The resulting fields are reconstructed from posed RGB captures of real scenes and support novel-view synthesis, semantic and free-space queries within a single grounded representation. This enables the efficient generation of diverse real-world environments to train and evaluate spatial reasoning models. As an example application, we outline an SRF-driven simulator for an orchard apple-reaching task, in which the radiance field supplies camera rendering, semantic ground truth, and occupancy queries to a physics engine.

cs.RO↗

arXiv2026

Multi-Agent Robotic Control with Onboard Vision-Language Models

Vision Language Models (VLMs) and Vision Language Action (VLA) models have shown promise in robotic control. Yet, they face significant challenges regarding explainability, generalization, and compute requirements. This paper presents a Multi-Agent System (MAS) architecture that addresses these limitations by deploying specialized agents on onboard hardware - eliminating dependence on external compute. The system controls a multi-purpose autonomous mobile manipulator in a simulated industrial warehouse, fulfilling five task categories: safety inspection, warehouse maintenance, warehouse search, package quality verification, and responding to human requests. Compact VLMs (3-20B parameters) are used throughout, with fine-tuning applied to improve package inspection accuracy. A novel "Megamind" orchestration agent mitigates context retention issues inherent to long-horizon planning with smaller models. The system was validated in a hardware-in-the-loop simulation using an AMD Ryzen(TM) AI mini PC. Results demonstrate that a fully onboard MAS architecture is a viable, cost-efficient alternative to cloud-dependent deployments, with strong potential for real-world transfer. The simulation environment has been released as open source under the Apache 2.0 licence.

cs.MA↗

arXiv2026

mllm-shap: A Shapley Value Explainability Platform for Text-Audio Multimodal Large Language Models

We introduce mllm-shap, an open-source Python framework designed to extend Shapley Value (SV) explainability from text-only Large Language Models to Multimodal LLMs (MLLMs) processing joint text and audio inputs. While text-based attribution is well-studied, mllm-shap addresses three critical challenges unique to the multimodal regime: (1) Modality-aware coalition masking, which manages the interleaved processing of discrete text tokens and dense audio encoder frames. (2) Multi-turn conversation tracking, utilizing per-token metadata to maintain role and modality context. (3) Phonetic alignment-based token grouping, a novel technique that reduces the coalition space by 10x to 50x, rendering SV estimation computationally feasible for long-form audio. The platform implements five SV estimation strategies, including a Complementary Contributions (CC) estimator with Neyman-optimal allocation that demonstrates superior convergence over standard Monte Carlo baselines. mllm-shap is provided as a pip-installable package featuring an interactive web-based GUI for granular attribution visualization. To our knowledge, this is the first publicly available framework providing a complete, reproducible pipeline for SV-based explainability in text-audio MLLMs.

cs.CL↗

arXiv2026

Bridging Traditional Explainability Methods and Multimodal Multilingual Models: An XAI-Based Analysis

Multimodal Large Language Models (MLLMs) effectively integrate text and audio to interpret context in complex interactive dialogues. However, the internal mechanisms by which heterogeneous modalities influence model behavior remain opaque. While Shapley Values (SV) provide a robust, model-agnostic framework for local explainability in text-based NLP, their extension to multimodal data is hindered by cross-channel dependencies, intricate dialogue structures, and the prohibitive computational complexity of dense audio representations. In this work, we formalize a multimodal extension of the Shapley Value framework, treating discrete text tokens and aligned audio segments as cooperative features. To ensure computational feasibility, we deploy a suite of efficient estimation strategies: exact SV computation for low-dimensional inputs and sampling-based approximations - including Monte Carlo permutations and stratified sampling with Neyman-optimal allocation - to minimize variance under constrained computational budgets. To resolve the granularity mismatch between modalities, we propose Spectrogram-Guided Phonetic Alignment (SGPA), a novel preprocessing method that maps high-frequency audio streams to interpretable, word-aligned segments. Our contribution is twofold: first, we provide an open-source, model-agnostic Python package and a companion GUI for the computation and interactive visualization of multimodal attributions. Second, we evaluate our framework using curated subsets of the VoiceBench and Infinity Instruct datasets across diverse multilingual scenarios. Our experimental results reveal that input modality is a primary driver of attribution volatility and demonstrate that standard syntactic importance proxies often fail to predict model attention in multimodal, cross-lingual contexts.

cs.CL↗

arXiv2026

Graph Neural Networks for Misinformation Detection: Performance-Efficiency Trade-offs

The rapid spread of online misinformation has led to increasingly complex detection models, including large language models and hybrid architectures. However, their computational cost and deployment limitations raise concerns about practical applicability. In this work, we benchmark graph neural networks (GNNs) against non-graph-based machine learning methods under controlled and comparable conditions. We evaluate lightweight GNN architectures (GCN, GraphSAGE, GAT, ChebNet) against Logistic Regression, Support Vector Machines, and Multilayer Perceptrons across seven public datasets in English, Indonesian, and Polish. All models use identical TF-IDF features to isolate the impact of relational structure. Performance is measured using F1 score, with inference time reported to assess efficiency. GNNs consistently outperform non-graph baselines across all datasets. For example, GraphSAGE achieves 96.8% F1 on Kaggle and 91.9% on WELFake, compared to 73.2% and 66.8% for MLP, respectively. On COVID-19, GraphSAGE reaches 90.5% F1 vs. 74.9%, while ChebNet attains 79.1% vs. 66.4% on FakeNewsNet. These gains are achieved with comparable or lower inference times. Overall, the results show that classic GNNs remain effective and efficient, challenging the need for increasingly complex architectures in misinformation detection.

cs.CL↗

arXiv2026

Clickbait detection: quick inference with maximum impact

We propose a lightweight hybrid approach to clickbait detection that combines OpenAI semantic embeddings with six compact heuristic features capturing stylistic and informational cues. To improve efficiency, embeddings are reduced using PCA and evaluated with XGBoost, GraphSAGE, and GCN classifiers. While the simplified feature design yields slightly lower F1-scores, graph-based models achieve competitive performance with substantially reduced inference time. High ROC--AUC values further indicate strong discrimination capability, supporting reliable detection of clickbait headlines under varying decision thresholds.

cs.CL↗

arXiv2026

SGPA: Spectrogram-Guided Phonetic Alignment for Feasible Shapley Value Explanations in Multimodal Large Language Models

Explaining the behavior of end-to-end audio language models via Shapley value attribution is intractable under native tokenization: a typical utterance yields over $150$ encoder frames, inflating the coalition space by roughly $10^{42}$ relative to text; individual audio frames lack standalone meaning; and token boundaries that bisect phonetic transitions introduce masking artifacts. We introduce Spectrogram-Guided Phonetic Alignment (SGPA), a four-stage pipeline that combines Connectionist Temporal Classification forced alignment with spectral boundary refinement to produce acoustically stable, word-aligned audio segments. Controlled diagnostics on LFM2-Audio-1.5B with VoiceBench show that SGPA yields a 43$\times$ reduction in model evaluations. Statistical testing confirms that SGPA significantly alters attribution concentration while preserving the global cumulative profile, establishing it as a feasibility-enabling layer for audio explainability.

cs.SD↗

arXiv2025

EnergyTwin: A Multi-Agent System for Simulating and Coordinating Energy Microgrids

Microgrids are deployed to reduce purchased grid energy, limit exposure to volatile tariffs, and ensure service continuity during disturbances. This requires coordinating heterogeneous distributed energy resources across multiple time scales and under variable conditions. Among existing tools, typically, power-system simulators capture physical behaviour but assume centralized control, while multi-agent frameworks model decentralized decision-making but represent energy with no physical grounding. In this context, the EnergyTwin is introduced, an agent-based microgrid simulation environment that couples physically grounded models with forecast-informed, rolling-horizon planning, and negotiations. Each asset is modeled as an agent, interacting with a central agent that obtains forecasts, formulates predictions, and allocates energy through contract-based interactions. EnergyTwin targets tertiary-layer decision making and is extensible for digital-twin use. Its feasibility was evaluated in a university campus microgrid scenario where multiple planning strategies were compared. Achieved results show that forecast-driven rolling-horizon planning increases local energy self-sufficiency, maintains higher battery reserves, and reduces exposure to low-resilience operating states. They demonstrate also potential of EnergyTwin as platform supporting research on resilient, negotiation-driven microgrids.

cs.MA↗

arXiv2025

Contemporary Agent Technology: LLM-Driven Advancements vs Classic Multi-Agent Systems

This contribution provides our comprehensive reflection on the contemporary agent technology, with a particular focus on the advancements driven by Large Language Models (LLM) vs classic Multi-Agent Systems (MAS). It delves into the models, approaches, and characteristics that define these new systems. The paper emphasizes the critical analysis of how the recent developments relate to the foundational MAS, as articulated in the core academic literature. Finally, it identifies key challenges and promising future directions in this rapidly evolving domain.

cs.MA↗

arXiv2025

RAI: Flexible Agent Framework for Embodied AI

With an increase in the capabilities of generative language models, a growing interest in embodied AI has followed. This contribution introduces RAI - a framework for creating embodied Multi Agent Systems for robotics. The proposed framework implements tools for Agents' integration with robotic stacks, Large Language Models, and simulations. It provides out-of-the-box integration with state-of-the-art systems like ROS 2. It also comes with dedicated mechanisms for the embodiment of Agents. These mechanisms have been tested on a physical robot, Husarion ROSBot XL, which was coupled with its digital twin, for rapid prototyping. Furthermore, these mechanisms have been deployed in two simulations: (1) robot arm manipulator and (2) tractor controller. All of these deployments have been evaluated in terms of their control capabilities, effectiveness of embodiment, and perception ability. The proposed framework has been used successfully to build systems with multiple agents. It has demonstrated effectiveness in all the aforementioned tasks. It also enabled identifying and addressing the shortcomings of the generative models used for embodied AI.

cs.MA↗

arXiv2025

Representing and querying data tensors in RDF and SPARQL

Embedding tensors in databases has recently gained in significance, due to the rapid proliferation of machine learning methods (including LLMs) which produce embeddings in the form of tensors. To support emerging use cases hybridizing machine learning with knowledge graphs, a robust and efficient tensor representation scheme is needed. We introduce a novel approach for representing data tensors as literals in RDF, along with an extension of SPARQL implementing specialized functionalities for handling such literals. The extension includes 36 SPARQL functions and four aggregates. To support this approach, we provide a thoroughly tested, open-source implementation based on Apache Jena, along with an exemplary knowledge graph and query set.

cs.DB↗

arXiv2025

ReReLRP -- Remembering and Recognizing Tasks with LRP

Deep neural networks have revolutionized numerous research fields and applications. Despite their widespread success, a fundamental limitation known as catastrophic forgetting remains, where models fail to retain their ability to perform previously learned tasks after being trained on new ones. This limitation is particularly acute in certain continual learning scenarios, where models must integrate the knowledge from new domains with their existing capabilities. Traditional approaches to mitigate this problem typically rely on memory replay mechanisms, storing either original data samples, prototypes, or activation patterns. Although effective, these methods often introduce significant computational overhead, raise privacy concerns, and require the use of dedicated architectures. In this work we present ReReLRP (Remembering and Recognizing with LRP), a novel solution that leverages Layerwise Relevance Propagation (LRP) to preserve information across tasks. Our contribution provides increased privacy of existing replay-free methods while additionally offering built-in explainability, flexibility of model architecture and deployment, and a new mechanism to increase memory storage efficiency. We validate our approach on a wide variety of datasets, demonstrating results comparable with a well-known replay-based method in selected scenarios.

cs.LG↗

arXiv2025

Applying Text Mining to Analyze Human Question Asking in Creativity Research

Creativity relates to the ability to generate novel and effective ideas in the areas of interest. How are such creative ideas generated? One possible mechanism that supports creative ideation and is gaining increased empirical attention is by asking questions. Question asking is a likely cognitive mechanism that allows defining problems, facilitating creative problem solving. However, much is unknown about the exact role of questions in creativity. This work presents an attempt to apply text mining methods to measure the cognitive potential of questions, taking into account, among others, (a) question type, (b) question complexity, and (c) the content of the answer. This contribution summarizes the history of question mining as a part of creativity research, along with the natural language processing methods deemed useful or helpful in the study. In addition, a novel approach is proposed, implemented, and applied to five datasets. The experimental results obtained are comprehensively analyzed, suggesting that natural language processing has a role to play in creative research.

cs.CL↗

arXiv2024

Overview of Current Challenges in Multi-Architecture Software Engineering and a Vision for the Future

The landscape of computing technologies is changing rapidly, straining existing software engineering practices and tools. The growing need to produce and maintain increasingly complex multi-architecture applications makes it crucial to effectively accelerate and automate software engineering processes. At the same time, artificial intelligence (AI) tools are expected to work hand-in-hand with human developers. Therefore, it becomes critical to model the software accurately, so that the AI and humans can share a common understanding of the problem. In this contribution, firstly, an in-depth overview of these interconnected challenges faced by modern software engineering is presented. Secondly, to tackle them, a novel architecture based on the emerging WebAssembly technology and the latest advancements in neuro-symbolic AI, autonomy, and knowledge graphs is proposed. The presented system architecture is based on the concept of dynamic, knowledge graph-based WebAssembly Twins, which model the software throughout all stages of its lifecycle. The resulting systems are to possess advanced autonomous capabilities, with full transparency and controllability by the end user. The concept takes a leap beyond the current software engineering approaches, addressing some of the most urgent issues in the field. Finally, the efforts towards realizing the proposed approach as well as future research directions are summarized.

cs.SE↗

arXiv2024

Realizing a Collaborative RDF Benchmark Suite in Practice

Collaborative mechanisms allow benchmarks to be updated continuously and adjust to the changing requirements and new use cases. This paradigm is employed for example in the field of machine learning, but up until now there were no examples of truly open and collaborative benchmarks for RDF systems. In this demo paper we present the collaboration functionalities of RiverBench, an open, multi-task RDF benchmark suite. Owing to its fully open and community-driven design, RiverBench allows any researcher or practitioner to submit a new dataset or benchmark task, report performed benchmark runs, and edit any resource in the suite. RiverBench's collaboration system is itself based on RDF and Linked Data mechanisms, and every resource in the suite has machine-readable RDF metadata. The showcased functionalities together make up a first-of-a-kind fully open and collaborative RDF benchmark suite. These features are meant to encourage other researchers to contribute to RiverBench, and make it a long-term project sustained by the community.

cs.DB↗

arXiv2024

Adaptive Active Inference Agents for Heterogeneous and Lifelong Federated Learning

Handling heterogeneity and unpredictability are two core problems in pervasive computing. The challenge is to seamlessly integrate devices with varying computational resources in a dynamic environment to form a cohesive system that can fulfill the needs of all participants. Existing work on adaptive systems typically focuses on optimizing individual variables or low-level Service Level Objectives (SLOs), such as constraining the usage of specific resources. While low-level control mechanisms permit fine-grained control over a system, they introduce considerable complexity, particularly in dynamic environments. To this end, we propose drawing from Active Inference (AIF), a neuroscientific framework for designing adaptive agents. Specifically, we introduce a conceptual agent for heterogeneous pervasive systems that permits setting global systems constraints as high-level SLOs. Instead of manually setting low-level SLOs, the system finds an equilibrium that can adapt to environmental changes. We demonstrate the viability of our AIF agents with an extensive experiment design, using heterogeneous and lifelong federated learning as an application scenario. We conduct our experiments on a physical testbed of devices with different resource types and vendor specifications. The results provide convincing evidence that an AIF agent can adapt a system to environmental changes. In particular, the AIF agent can balance competing SLOs in resource heterogeneous environments to ensure up to 98% fulfillment rate.

cs.LG↗

arXiv2024

Fake News Detection: It's All in the Data!

This comprehensive survey serves as an indispensable resource for researchers embarking on the journey of fake news detection. By highlighting the pivotal role of dataset quality and diversity, it underscores the significance of these elements in the effectiveness and robustness of detection models. The survey meticulously outlines the key features of datasets, various labeling systems employed, and prevalent biases that can impact model performance. Additionally, it addresses critical ethical issues and best practices, offering a thorough overview of the current state of available datasets. Our contribution to this field is further enriched by the provision of GitHub repository, which consolidates publicly accessible datasets into a single, user-friendly portal. This repository is designed to facilitate and stimulate further research and development efforts aimed at combating the pervasive issue of fake news.

cs.CL↗

arXiv2024

Not All RDF is Created Equal: Investigating RDF Load Times on Resource-Constrained Devices

As the role of knowledge-based systems in IoT keeps growing, ensuring resource efficiency of RDF stores becomes critical. However, up until now benchmarks of RDF stores were most often conducted with only one dataset, and the differences between the datasets were not explored in detail. In this paper, our objective is to close this research gap by experimentally evaluating the load times of eight diverse RDF datasets from the RiverBench benchmark suite. In the experiments, we use five different RDF store implementations and several resource-constrained hardware platforms. To analyze the results, we introduce the notion of relative loading speed (RLS), allowing us to observe that the loading speed can differ between datasets by as much as a factor of 9.01. This serves as clear evidence that ``not all RDF is created equal'' and stresses the importance of using multiple benchmark datasets in evaluations. We outline the possible reasons for this drastic difference, which should be further investigated in future work. To this end, we published the data, code, and the results of our experiments.

cs.DB↗