9 papers
CycleCap: Improving VLMs Captioning Performance via Self-Supervised Cycle Consistency Fine-Tuning
Marios Krestenitis, Christos Tzelepis, Konstantinos Ioannidis +5
Visual-Language Models (VLMs) have achieved remarkable progress in image captioning, visual question answering, and visual reasoning. Yet they remain prone to vision-language misal…
Understanding the Performance Plateau in Text-to-Video Retrieval: A Comprehensive Empirical and Linguistic Analysis
Maria-Eirini Pegia, Dimitrios Stefanopoulos, Björn Ãór Jónsson +4
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…
Utilizing Large Language Models for Machine Learning Explainability
Alexandros Vassiliades, Nikolaos Polatidis, Stamatios Samaras +5
This study explores the explainability capabilities of large language models (LLMs), when employed to autonomously generate machine learning (ML) solutions. We examine two classifi…
Frugal Machine Learning for Energy-efficient, and Resource-aware Artificial Intelligence
John Violos, Konstantina-Christina Diamanti, Ioannis Kompatsiaris +1
Frugal Machine Learning (FML) refers to the practice of designing Machine Learning (ML) models that are efficient, cost-effective, and mindful of resource constraints. This field a…
Few-Shot Class-Incremental Learning For Efficient SAR Automatic Target Recognition
George Karantaidis, Athanasios Pantsios, Ioannis Kompatsiaris +1
Synthetic aperture radar automatic target recognition (SAR-ATR) systems have rapidly evolved to tackle incremental recognition challenges in operational settings. Data scarcity rem…
A Brief Review for Compression and Transfer Learning Techniques in DeepFake Detection
Andreas Karathanasis, John Violos, Ioannis Kompatsiaris +1
Training and deploying deepfake detection models on edge devices offers the advantage of maintaining data privacy and confidentiality by processing it close to its source. However,…