22 citations · 108 across the 43 of their papers we have counts for
46 papers
Attention-Aware Routing: Coupling Routing and Attention in MoEs
Despoina Kosmopoulou, Anastasios Tsetsilas, Efthymios Georgiou +3
In Mixture-of-Experts language models, the router typically selects and weights experts based on the token's hidden state, utilizing limited contextual information. We propose Atte…
The Illusion of Balanced Multimodal Sentiment Analysis: Beyond the Limits of Optimization-Based Methods
Ioanna Kaffeza, Efthymios Georgiou, Alexandros Potamianos
Multimodal Sentiment Analysis (MSA) remains constrained by modality imbalance, yet the field continues to rely on optimization-based balancing methods that promise more than they d…
Deterministic Prompting for Speaker-Stable Low-Resource Greek TTS
Georgios Syllas, Efthymios Georgiou, Kosmas Kritsis +1
Modern TTS systems approach human quality for high-resource languages but degrade when clean speech data is scarce. Modern Greek exemplifies this, lacking the curated corpora behin…
Cognitive Profiling of LRMs' Reasoning Traces Using Bloom's Taxonomy
Maria-Eleni Zoumpoulidi, Georgios Paraskevopoulos, Alexandros Potamianos
Large Reasoning Models (LRMs) have revolutionized reasoning in LLMs, and the increasing public availability of reasoning traces creates valuable opportunities to study model behavi…
Segregate, Refine, Integrate: Decomposing Multimodal Fusion for Sentiment Analysis
Alexios Filippakopoulos, Elias Kallioras, Nikolaos Xiros +2
Multimodal fusion must simultaneously refine modality-specific signals and model cross-modal interactions; two competing objectives typically entangled within the same operation. W…
Semantic F1 Scores: Fair Evaluation Under Fuzzy Class Boundaries
Georgios Chochlakis, Jackson Trager, Vedant Jhaveri +3
We propose Semantic F1 Scores, novel evaluation metrics for subjective or fuzzy multi-label classification that quantify semantic relatedness between predicted and gold labels. Unl…
Masked Diffusion Language Models with Frequency-Informed Training
Despoina Kosmopoulou, Efthymios Georgiou, Vaggelis Dorovatas +2
We present a masked diffusion language modeling framework for data-efficient training for the BabyLM 2025 Challenge. Our approach applies diffusion training objectives to language…
MSDA: Combining Pseudo-labeling and Self-Supervision for Unsupervised Domain Adaptation in ASR
Dimitrios Damianos, Georgios Paraskevopoulos, Alexandros Potamianos
In this work, we investigate the Meta PL unsupervised domain adaptation framework for Automatic Speech Recognition (ASR). We introduce a Multi-Stage Domain Adaptation pipeline (MSD…
Auto-Compressing Networks
Vaggelis Dorovatas, Georgios Paraskevopoulos, Alexandros Potamianos
Deep neural networks with short residual connections have demonstrated remarkable success across domains, but increasing depth often introduces computational redundancy without cor…
MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions
Georgios Chatzichristodoulou, Despoina Kosmopoulou, Antonios Kritikos +5
SER is a challenging task due to the subjective nature of human emotions and their uneven representation under naturalistic conditions. We propose MEDUSA, a multimodal framework wi…
DeepMLF: Multimodal language model with learnable tokens for deep fusion in sentiment analysis
Efthymios Georgiou, Vassilis Katsouros, Yannis Avrithis +1
While multimodal fusion has been extensively studied in Multimodal Sentiment Analysis (MSA), the role of fusion depth and multimodal capacity allocation remains underexplored. In t…
Aggregation Artifacts in Subjective Tasks Collapse Large Language Models' Posteriors
Georgios Chochlakis, Alexandros Potamianos, Kristina Lerman +1
In-context Learning (ICL) has become the primary method for performing natural language tasks with Large Language Models (LLMs). The knowledge acquired during pre-training is cruci…
BloomWise: Enhancing Problem-Solving capabilities of Large Language Models using Bloom's-Taxonomy-Inspired Prompts
Maria-Eleni Zoumpoulidi, Georgios Paraskevopoulos, Alexandros Potamianos
Despite the remarkable capabilities of large language models (LLMs) across a range of tasks, mathematical reasoning remains a challenging frontier. Motivated by the observation tha…
Y-Drop: A Conductance based Dropout for fully connected layers
Efthymios Georgiou, Georgios Paraskevopoulos, Alexandros Potamianos
In this work, we introduce Y-Drop, a regularization method that biases the dropout algorithm towards dropping more important neurons with higher probability. The backbone of our ap…
The Strong Pull of Prior Knowledge in Large Language Models and Its Impact on Emotion Recognition
Georgios Chochlakis, Alexandros Potamianos, Kristina Lerman +1
In-context Learning (ICL) has emerged as a powerful paradigm for performing natural language tasks with Large Language Models (LLM) without updating the models' parameters, in cont…
PowMix: A Versatile Regularizer for Multimodal Sentiment Analysis
Efthymios Georgiou, Yannis Avrithis, Alexandros Potamianos
Multimodal sentiment analysis (MSA) leverages heterogeneous data sources to interpret the complex nature of human sentiments. Despite significant progress in multimodal architectur…
From West to East: Who can understand the music of the others better?
Charilaos Papaioannou, Emmanouil Benetos, Alexandros Potamianos
Recent developments in MIR have led to several benchmark deep learning models whose embeddings can be used for a variety of downstream tasks. At the same time, the vast majority of…
Self-Attention Based Generative Adversarial Networks For Unsupervised Video Summarization
Maria Nektaria Minaidi, Charilaos Papaioannou, Alexandros Potamianos
In this paper, we study the problem of producing a comprehensive video summary following an unsupervised approach that relies on adversarial learning. We build on a popular method…
SeqAug: Sequential Feature Resampling as a modality agnostic augmentation method
Efthymios Georgiou, Alexandros Potamianos
Data augmentation is a prevalent technique for improving performance in various machine learning applications. We propose SeqAug, a modality-agnostic augmentation method that is ta…
Efficient Audio Captioning Transformer with Patchout and Text Guidance
Thodoris Kouzelis, Grigoris Bastas, Athanasios Katsamanis +1
Automated audio captioning is multi-modal translation task that aim to generate textual descriptions for a given audio clip. In this paper we propose a full Transformer architectur…
Depression detection in social media posts using affective and social norm features
Ilias Triantafyllopoulos, Georgios Paraskevopoulos, Alexandros Potamianos
We propose a deep architecture for depression detection from social media posts. The proposed architecture builds upon BERT to extract language representations from social media po…
Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems A case study for Modern Greek
Georgios Paraskevopoulos, Theodoros Kouzelis, Georgios Rouvalis +3
Modern speech recognition systems exhibits rapid performance degradation under domain shift. This issue is especially prevalent in data-scarce settings, such as low-resource langua…
Alternating Objectives Generates Stronger PGD-Based Adversarial Attacks
Nikolaos Antoniou, Efthymios Georgiou, Alexandros Potamianos
Designing powerful adversarial attacks is of paramount importance for the evaluation of -bounded adversarial defenses. Projected Gradient Descent (PGD) is one of the most e…
Adapted Multimodal BERT with Layer-wise Fusion for Sentiment Analysis
Odysseas S. Chlapanis, Georgios Paraskevopoulos, Alexandros Potamianos
Multimodal learning pipelines have benefited from the success of pretrained language models. However, this comes at the cost of increased model parameters. In this work, we propose…