arXiv subjects

Stefanos Vrochidis

Publications and source records attributed to Stefanos Vrochidis.

11 recordsLinked to original sources

arXiv2026

ACTING: A Platform for Cyber Ranges Federation

Cyber Defence (CD) training requires interoperable cyber-range environments capable of supporting complex, multidomain exercises across distributed infrastructures. This paper presents three main contributions addressing this challenge. First, we introduce the Exercise Description Language - First Generation (EDL-FG), a structured language for formally describing cyber-range training services and exercises. EDL-FG captures both the technical infrastructure required to emulate ICT/OT environments and the scenario logic governing cyber events, injects, and participant interactions, enabling interoperable and automated scenario deployment across federated Cyber Ranges (CRs). Second, the ACTING platform introduces automated PE and scoring mechanisms that assess trainee actions during exercises through coordinated data collection and analysis across participating CRs. Third, the platform enables multi-domain cyber training scenarios that combine civilian and military operational contexts. Building upon federation capabilities established under the H2020 ECHO project, ACTING demonstrates how interoperable scenario description and automated evaluation support scalable and realistic CD training.

cs.CR↗

arXiv2026

CycleCap: Improving VLMs Captioning Performance via Self-Supervised Cycle Consistency Fine-Tuning

Visual-Language Models (VLMs) have achieved remarkable progress in image captioning, visual question answering, and visual reasoning. Yet they remain prone to vision-language misalignment, often producing overly generic or hallucinated descriptions. Existing approaches address this via instruction tuning-requiring costly, large-scale annotated datasets or via complex test-time frameworks for caption refinement. In this work, we revisit image-text alignment through the lens of cycle consistency: given an image and a caption generated by an image-to-text model, the backward mapping through a text-to-image model should reconstruct an image that closely matches the original. In our setup, a VLM serves as the image-to-text component, while a pre-trained text-to-image model closes the loop by reconstructing the image from the generated caption. Building on this, we introduce CycleCap, a fine-tuning scheme to improve image captioning using Group Relative Policy Optimization (GRPO) with a reward based on the similarity between the original and reconstructed images, computed on-the-fly. Unlike previous work that uses cycle consistency loss for preference dataset construction, our method leverages cycle consistency directly as a self-supervised training signal. This enables the use of raw images alone, eliminating the need for curated image-text datasets, while steering the VLM to produce more accurate and grounded text descriptions. Applied to four VLMs ranging from 1B to 7B parameters, CycleCap yields consistent improvements across captioning and hallucination benchmarks, surpassing state-of-the-art methods that rely on supervised cycle consistency training.

cs.CV↗

arXiv2026

Understanding the Performance Plateau in Text-to-Video Retrieval: A Comprehensive Empirical and Linguistic Analysis

Text-to-video retrieval enables users to find relevant video content using natural language queries, a task that has grown increasingly important with the rapid expansion of online video. Over the past six years, research has produced numerous methods, such as dual encoders, attention-driven models, and multimodal fusion approaches; however, fundamental questions remain about model behavior, dataset influence, and query difficulty. In this work, we evaluate 14 state-of-the-art retrieval methods across 3 widely used datasets under a unified preprocessing and evaluation framework. We analyze caption characteristics, including length, clarity, semantic category, and Action vs. Scene balance, and link these to model performance. Our results show that short, clear, and simple captions, such as those describing single actions or color attributes, achieve higher recall, while complex events, multi-step activities, or fine-grained scene descriptions remain challenging for all existing models. Attention-driven architectures better handle temporally dependent or multi-step queries, whereas dual-encoder and multimodal fusion models perform well primarily on simpler or single-category captions. Cross-dataset generalization improves with larger, more diverse caption sets, but generative captions do not consistently enhance retrieval accuracy. Overall, our findings highlight key dataset factors, benchmark challenges, and the interplay between query content and model architecture, providing guidance for developing more effective text-to-video retrieval systems.

cs.IR↗

arXiv2025

Utilizing Large Language Models for Machine Learning Explainability

This study explores the explainability capabilities of large language models (LLMs), when employed to autonomously generate machine learning (ML) solutions. We examine two classification tasks: (i) a binary classification problem focused on predicting driver alertness states, and (ii) a multilabel classification problem based on the yeast dataset. Three state-of-the-art LLMs (i.e. OpenAI GPT, Anthropic Claude, and DeepSeek) are prompted to design training pipelines for four common classifiers: Random Forest, XGBoost, Multilayer Perceptron, and Long Short-Term Memory networks. The generated models are evaluated in terms of predictive performance (recall, precision, and F1-score) and explainability using SHAP (SHapley Additive exPlanations). Specifically, we measure Average SHAP Fidelity (Mean Squared Error between SHAP approximations and model outputs) and Average SHAP Sparsity (number of features deemed influential). The results reveal that LLMs are capable of producing effective and interpretable models, achieving high fidelity and consistent sparsity, highlighting their potential as automated tools for interpretable ML pipeline generation. The results show that LLMs can produce effective, interpretable pipelines with high fidelity and consistent sparsity, closely matching manually engineered baselines.

cs.LG↗

arXiv2024

RAG Playground: A Framework for Systematic Evaluation of Retrieval Strategies and Prompt Engineering in RAG Systems

We present RAG Playground, an open-source framework for systematic evaluation of Retrieval-Augmented Generation (RAG) systems. The framework implements and compares three retrieval approaches: naive vector search, reranking, and hybrid vector-keyword search, combined with ReAct agents using different prompting strategies. We introduce a comprehensive evaluation framework with novel metrics and provide empirical results comparing different language models (Llama 3.1 and Qwen 2.5) across various retrieval configurations. Our experiments demonstrate significant performance improvements through hybrid search methods and structured self-evaluation prompting, achieving up to 72.7% pass rate on our multi-metric evaluation framework. The results also highlight the importance of prompt engineering in RAG systems, with our custom-prompted agents showing consistent improvements in retrieval accuracy and response quality.

cs.LG↗

arXiv2024

A Multi-Task Text Classification Pipeline with Natural Language Explanations: A User-Centric Evaluation in Sentiment Analysis and Offensive Language Identification in Greek Tweets

Interpretability is a topic that has been in the spotlight for the past few years. Most existing interpretability techniques produce interpretations in the form of rules or feature importance. These interpretations, while informative, may be harder to understand for non-expert users and therefore, cannot always be considered as adequate explanations. To that end, explanations in natural language are often preferred, as they are easier to comprehend and also more presentable to end-users. This work introduces an early concept for a novel pipeline that can be used in text classification tasks, offering predictions and explanations in natural language. It comprises of two models: a classifier for labelling the text and an explanation generator which provides the explanation. The proposed pipeline can be adopted by any text classification task, given that ground truth rationales are available to train the explanation generator. Our experiments are centred around the tasks of sentiment analysis and offensive language identification in Greek tweets, using a Greek Large Language Model (LLM) to obtain the necessary explanations that can act as rationales. The experimental evaluation was performed through a user study based on three different metrics and achieved promising results for both datasets.

cs.CL↗

arXiv2023

Overcome the Fear Of Missing Out: Active Sensing UAV Scanning for Precision Agriculture

This paper deals with the problem of informative path planning for a UAV deployed for precision agriculture applications. First, we observe that the ``fear of missing out'' data lead to uniform, conservative scanning policies over the whole agricultural field. Consequently, employing a non-uniform scanning approach can mitigate the expenditure of time in areas with minimal or negligible real value, while ensuring heightened precision in information-dense regions. Turning to the available informative path planning methodologies, we discern that certain methods entail intensive computational requirements, while others necessitate training on an ideal world simulator. To address the aforementioned issues, we propose an active sensing coverage path planning approach, named OverFOMO, that regulates the speed of the UAV in accordance with both the relative quantity of the identified classes, i.e. crops and weeds, and the confidence level of such detections. To identify these instances, a robust Deep Learning segmentation model is deployed. The computational needs of the proposed algorithm are independent of the size of the agricultural field, rendering its applicability on modern UAVs quite straightforward. The proposed algorithm was evaluated with a simu-realistic pipeline, combining data from real UAV missions and the high-fidelity dynamics of AirSim simulator, showcasing its performance improvements over the established state of affairs for this type of missions. An open-source implementation of the algorithm and the evaluation pipeline is also available: \url{https://github.com/emmarapt/OverFOMO}.

cs.RO↗

arXiv2023

Neural Crystals

We face up to the challenge of explainability in Multimodal Artificial Intelligence (MMAI). At the nexus of neuroscience-inspired and quantum computing, interpretable and transparent spin-geometrical neural architectures for early fusion of large-scale, heterogeneous, graph-structured data are envisioned, harnessing recent evidence for relativistic quantum neural coding of (co-)behavioral states in the self-organizing brain, under competitive, multidimensional dynamics. The designs draw on a self-dual classical description - via special Clifford-Lipschitz operations - of spinorial quantum states within registers of at most 16 qubits for efficient encoding of exponentially large neural structures. Formally 'trained', Lorentz neural architectures with precisely one lateral layer of exclusively inhibitory interneurons accounting for anti-modalities, as well as their co-architectures with intra-layer connections are highlighted. The approach accommodates the fusion of up to 16 time-invariant interconnected (anti-)modalities and the crystallization of latent multidimensional patterns. Comprehensive insights are expected to be gained through applications to Multimodal Big Data, under diverse real-world scenarios.

q-bio.NC↗

arXiv2023

User Identity Linkage in Social Media Using Linguistic and Social Interaction Features

Social media users often hold several accounts in their effort to multiply the spread of their thoughts, ideas, and viewpoints. In the particular case of objectionable content, users tend to create multiple accounts to bypass the combating measures enforced by social media platforms and thus retain their online identity even if some of their accounts are suspended. User identity linkage aims to reveal social media accounts likely to belong to the same natural person so as to prevent the spread of abusive/illegal activities. To this end, this work proposes a machine learning-based detection model, which uses multiple attributes of users' online activity in order to identify whether two or more virtual identities belong to the same real natural person. The models efficacy is demonstrated on two cases on abusive and terrorism-related Twitter content.

cs.SI↗

arXiv2020

Towards Semantic Detection of Smells in Cloud Infrastructure Code

Automated deployment and management of Cloud applications relies on descriptions of their deployment topologies, often referred to as Infrastructure Code. As the complexity of applications and their deployment models increases, developers inadvertently introduce software smells to such code specifications, for instance, violations of good coding practices, modular structure, and more. This paper presents a knowledge-driven approach enabling developers to identify the aforementioned smells in deployment descriptions. We detect smells with SPARQL-based rules over pattern-based OWL 2 knowledge graphs capturing deployment models. We show the feasibility of our approach with a prototype and three case studies.

cs.SE↗

arXiv2016

Towards Air Quality Estimation Using Collected Multimodal Environmental Data

This paper presents an open platform, which collects multimodal environmental data related to air quality from several sources including official open sources, social media and citizens. Collecting and fusing different sources of air quality data into a unified air quality indicator is a highly challenging problem, leveraging recent advances in image analysis, open hardware, machine learning and data fusion and is expected to result in increased geographical coverage and temporal granularity of air quality data.

cs.CY↗