arXiv subjects

John Pavlopoulos

Publications and source records attributed to John Pavlopoulos.

At least 19 recordsLinked to original sources

arXiv2026

Aristotelian Virtue Profiling of LLMs through Ethical Dilemmas

Large Language Models (LLMs) often face ethical tradeoffs in which several responses may be defensible but express different priorities, such as fairness, honesty, courage, or restraint. We introduce VirtueMap, a framework for describing these patterns through an Aristotelian virtue-ethics lens. Instead of asking for a single correct answer, VirtueMap asks humans or LLMs to rank all five responses to each of seven general, non-lethal, non-political, and non-religious ethical dilemmas. To define the reference orderings used for scoring, we first proposed, for each dilemma and virtue, an ordering of the five responses from most to least expressive of that virtue. We then collected more than 100 respondent evaluations per ordering and retained it as operational ground truth only when at least 95% confirmed it. Rankings are scored against these retained orderings using normalized Borda alignment, yielding profiles over Practical Wisdom, Justice, Truthfulness, Courage, and Temperance. We apply VirtueMap to nine LLM families in a repeated-run evaluation and find high mean rank consistency (90.3%), with the largest differences appearing on Courage, Temperance, and Justice. We also release an interactive website that computes profiles locally in the browser and compares respondents with measured LLM profiles.

cs.AI↗

arXiv2026

Learning Diachronic Representations of Ancient Greek Letterforms

Learning representations that remain robust across centuries of variation in handwriting is a key challenge in diachronic representation learning. Taking one of the longest continuously used writing systems, ancient Greek, as a case study, we introduce three datasets for diachronic representation learning: Hell-Char, a curated training set spanning the 3rd-1st centuries BCE, and two evaluation sets, PaLit-Char (2nd-5th c. CE) and Med-Char (9th-14th c. CE). To address the challenges of symbolic variation, scarce data, and systematic degradation, we propose: a similarity-weighted supervised contrastive loss that biases embeddings using dynamically estimated inter-class similarities, and a lacuna-driven augmentation scheme that simulates realistic manuscript corruptions. Trained with these strategies, both a lightweight CNN and a pretrained ResNet achieve strong recognition performance and produce embeddings that more coherently separate character classes than PCA or generic pretrained models. These embeddings enable clustering, identification of stylistic subgroups, and construction of prototype images that visualize diachronic evolution and transitional letterforms. Our results demonstrate that respecting intrinsic inter-letter relationships and augmenting with domain-informed corruptions yield robust, interpretable representations, offering a transferable paradigm for representation learning under scarce, temporally evolving, and noisy conditions. Code and data available at: https://github.com/ipavlopoulos/diachronic-greek-letterforms.

cs.LG↗

arXiv2026

To Facilitate or not to Facilitate: Human and LLM Facilitator Tendencies in Online Discussions

Automating facilitation in online discussions is a long-standing social concern given the increasing time we spend on online spaces and the failure of content moderation approaches. While studies have been conducted on how to facilitate, none have answered the essential question of when to do so. A potential answer is using LLMs, which ostensibly make automated, large-scale intervention increasingly feasible. In this study, we examine when LLMs decide to facilitate by defining what facilitation is, observing when humans decide to facilitate, and comparing their decisions with those made by LLMs. To this end, we create PEFK, a corpus standardizing and aggregating all relevant facilitation datasets. We are the first to run a survey on facilitation timing, which we execute using expert facilitative participants and LLM-as-a-judge models. We discover that while humans are more cautious, LLMs are excessively eager to facilitate, although both are more certain when judging that facilitation is not needed. We then investigate whether this behavior can be corrected using alternative setups for LLMs and training ModernBert classifiers on established datasets, finding that the latter perform more reliably than the former, although current datasets impose a relatively low performance ceiling.

cs.HC↗

arXiv2026

Composite Silhouette: A Subsampling-based Aggregation Strategy

Determining the number of clusters is a central challenge in unsupervised learning, where ground-truth labels are unavailable. The Silhouette coefficient is a widely used internal validation metric for this task, yet its standard micro-averaged form tends to favor larger clusters under size imbalance. Macro-averaging mitigates this bias by weighting clusters equally, but may overemphasize noise from under-represented groups. We introduce Composite Silhouette, an internal criterion for cluster-count selection that aggregates evidence across repeated subsampled clusterings rather than relying on a single partition. For each subsample, micro- and macro-averaged Silhouette scores are combined through an adaptive convex weight determined by their normalized discrepancy and smoothed by a bounded nonlinearity; the final score is then obtained by averaging these subsample-level composites. We establish key properties of the criterion and derive finite-sample concentration guarantees for its subsampling estimate. Experiments on synthetic and real-world datasets show that Composite Silhouette effectively reconciles the strengths of micro- and macro-averaging, yielding more accurate recovery of the ground-truth number of clusters.

cs.LG↗

arXiv2026

CAKE: Confidence in Assignments via K-partition Ensembles

Clustering is widely used for unsupervised structure discovery, yet it offers limited insight into how reliable each individual assignment is. Diagnostics, such as convergence behavior or objective values, may reflect global quality, but they do not indicate whether particular instances are assigned confidently, especially for initialization-sensitive algorithms like k-means. This assignment-level instability can undermine both accuracy and robustness. Ensemble approaches improve global consistency by aggregating multiple runs, but they typically lack tools for quantifying pointwise confidence in a way that combines cross-run agreement with geometric support from the learned cluster structure. This work introduces CAKE (Confidence in Assignments via K-partition Ensembles), a framework that evaluates each point using two complementary statistics computed over a clustering ensemble: assignment stability and consistency of local geometric fit. These are combined into a single, interpretable score in [0,1]. The theoretical analysis shows that CAKE remains effective under noise and separates stable from unstable points. Experiments on synthetic and real-world datasets indicate that CAKE effectively highlights ambiguous points and stable core members, providing a confidence ranking over instances that can be used for selection or prioritization in downstream clustering workflows.

cs.LG↗

arXiv2026

Quantifying Retriever-Generator Alignment in RAG with Local Explanations

Retrieval-Augmented Generation (RAG) systems combine dense retrievers and language models to ground their outputs in external documents. However, the interaction between these components remains opaque, creating challenges for deployment in high-stakes domains. We present RAG-E, an end-to-end explainability framework that quantifies retriever-generator alignment through mathematically grounded attribution methods. Our approach adapts Integrated Gradients for retriever analysis, proposes a Monte Carlo-stabilized Shapley Value approximation for generator attribution, and introduces the Weighted Attribution-Relevance Gap (WARG) metric to measure how closely the generator's document usage aligns with retriever rankings. Experiments on PopQA, QAMPARI, and TREC CAST datasets reveal substantial misalignment: depending on the model and setting, generators often ignore top-ranked documents and rely on documents ranked as less relevant. We show that WARG captures retriever-generator alignment better than Pearson and Spearman correlations and can serve as an indicator of RAG performance. RAG-E and WARG provide a practical framework for auditing this interaction, enabling more reliable and transparent RAG systems.

cs.CL↗

arXiv2026

Are we chasing ghosts? Quantifying unattributable polarization, and attributing the rest to annotator groups

Standard agreement metrics often fail to capture systematic differences in opinion between minority and majority-group annotators, jeopardizing tasks such as hate speech and toxicity detection. Polarization has recently been proposed as a more robust way of distinguishing minor disagreements from systematic differences in opinion, but existing approaches do not provide practical tools for attributing it to specific annotator groups. We evaluate current methods and identify two major limitations in realistic settings: (1) the presence of ``inherent'' polarization that cannot be attributed to any known or latent groups, and (2) opposing polarization effects canceling each other out in aggregated annotations. To address these issues, we introduce a new metric that measures and tests the statistical significance of polarization attribution for annotator groups while avoiding these limitations, as well as an open-source Python library implementation, finding that no more than 20 annotators are needed per comment for reliable estimation. We apply our method to four subjective NLP datasets and find that gender and race consistently explain polarization patterns, while differences between annotator groups become stronger as the groups are further apart.

cs.CL↗

arXiv2025

Proverbs or Pythian Oracles? Sentiments and Emotions in Greek Sayings

Proverbs are among the most fascinating language phenomena that transcend cultural and linguistic boundaries. Yet, much of the global landscape of proverbs remains underexplored, as many cultures preserve their traditional wisdom within their own communities due to the oral tradition of the phenomenon. Taking advantage of the current advances in Natural Language Processing (NLP), we focus on Greek proverbs, analyzing their sentiment and emotion. Departing from an annotated dataset of Greek proverbs, (1) we propose a multi-label annotation framework and dataset that captures the emotional variability of the proverbs, (2) we up-scale to local varieties, (3) we sketch a map of Greece that provides an overview of the distribution of emotions. Our findings show that the interpretation of proverbs is multidimensional, a property manifested through both multi-labeling and instance-level polarity. LLMs can capture and reproduce this complexity, and can therefore help us better understand the proverbial landscape of a place, as in the case of Greece, where surprise and anger compete and coexist within proverbs.

cs.CL↗

arXiv2025

TopClustRAG at SIGIR 2025 LiveRAG Challenge

We present TopClustRAG, a retrieval-augmented generation (RAG) system developed for the LiveRAG Challenge, which evaluates end-to-end question answering over large-scale web corpora. Our system employs a hybrid retrieval strategy combining sparse and dense indices, followed by K-Means clustering to group semantically similar passages. Representative passages from each cluster are used to construct cluster-specific prompts for a large language model (LLM), generating intermediate answers that are filtered, reranked, and finally synthesized into a single, comprehensive response. This multi-stage pipeline enhances answer diversity, relevance, and faithfulness to retrieved evidence. Evaluated on the FineWeb Sample-10BT dataset, TopClustRAG ranked 2nd in faithfulness and 7th in correctness on the official leaderboard, demonstrating the effectiveness of clustering-based context filtering and prompt aggregation in large-scale RAG systems.

cs.CL↗

arXiv2025

Silhouette-Driven Instance-Weighted $k$-means

Clustering is a fundamental unsupervised learning task with applications across a wide range of domains. Popular algorithms such as $k$-means are efficient and widely used, but can be sensitive to outliers, ambiguous boundary points, and heterogeneous cluster geometry, which may distort centroid estimates and yield suboptimal partitions. We introduce K-Sil, a silhouette-driven $k$-means variant that, at each iteration, weights points using a centroid-margin proxy for the silhouette score, emphasizing confidently assigned instances while down-weighting borderline or noisy regions. Centroid updates take the form of a softmax-weighted mean, and an adaptive temperature automatically calibrates the sharpness of the weight distribution using a cluster-balanced, macro-averaged, silhouette criterion. Under standard separation conditions, we establish a local convergence result for the induced weighted centroid updates. Experiments on 15 real-world datasets spanning tabular, biomedical, text, and image representations show consistent gains in internal validation metrics and typical improvements in external validation metrics over $k$-means and competitive instance-weighted baselines.

cs.LG↗

arXiv2025

Learning to Align: Addressing Character Frequency Distribution Shifts in Handwritten Text Recognition

Handwritten text recognition aims to convert visual input into machine-readable text, and it remains challenging due to the evolving and context-dependent nature of handwriting. Character sets change over time, and character frequency distributions shift across historical periods or regions, often causing models trained on broad, heterogeneous corpora to underperform on specific subsets. To tackle this, we propose a novel loss function that incorporates the Wasserstein distance between the character frequency distribution of the predicted text and a target distribution empirically derived from training data. By penalizing divergence from expected distributions, our approach enhances both accuracy and robustness under temporal and contextual intra-dataset shifts. Furthermore, we demonstrate that character distribution alignment can also improve existing models at inference time without requiring retraining by integrating it as a scoring function in a guided decoding scheme. Experimental results across multiple datasets and architectures confirm the effectiveness of our method in boosting generalization and performance. We open source our code at https://github.com/pkaliosis/fada.

cs.CV↗

arXiv2025

Dialect Normalization using Large Language Models and Morphological Rules

Natural language understanding systems struggle with low-resource languages, including many dialects of high-resource ones. Dialect-to-standard normalization attempts to tackle this issue by transforming dialectal text so that it can be used by standard-language tools downstream. In this study, we tackle this task by introducing a new normalization method that combines rule-based linguistically informed transformations and large language models (LLMs) with targeted few-shot prompting, without requiring any parallel data. We implement our method for Greek dialects and apply it on a dataset of regional proverbs, evaluating the outputs using human annotators. We then use this dataset to conduct downstream experiments, finding that previous results regarding these proverbs relied solely on superficial linguistic information, including orthographic artifacts, while new observations can still be made through the remaining semantics.

cs.CL↗

arXiv2025

SemEval-2025 Task 9: The Food Hazard Detection Challenge

In this challenge, we explored text-based food hazard prediction with long tail distributed classes. The task was divided into two subtasks: (1) predicting whether a web text implies one of ten food-hazard categories and identifying the associated food category, and (2) providing a more fine-grained classification by assigning a specific label to both the hazard and the product. Our findings highlight that large language model-generated synthetic data can be highly effective for oversampling long-tail distributions. Furthermore, we find that fine-tuned encoder-only, encoder-decoder, and decoder-only systems achieve comparable maximum performance across both subtasks. During this challenge, we gradually released (under CC BY-NC-SA 4.0) a novel set of 6,644 manually labeled food-incident reports.

cs.CL↗

arXiv2025

Designing Synthetic Discussion Generation Systems: A Case Study for Online Facilitation

A critical challenge in social science research is the high cost associated with experiments involving human participants. We identify Synthetic Discussion Generation (SDG), a novel Natural Language Processing (NLP) direction aimed at creating simulated discussions that enable cost-effective pilot experiments and develop a theoretical, task-agnostic framework for designing, evaluating, and implementing these simulations. We argue that the use of proprietary models such as the OpenAI GPT family for such experiments is often unjustified in terms of both cost and capability, despite its prevalence in current research. Our experiments demonstrate that smaller quantized models (7B-8B) can produce effective simulations at a cost more than 44 times lower compared to their proprietary counterparts. We use our framework in the context of online facilitation, where humans actively engage in discussions to improve them, unlike more conventional content moderation. By treating this problem as a downstream task for our framework, we show that synthetic simulations can yield generalizable results at least by revealing limitations before engaging human discussants. In LLM facilitators, a critical limitation is that they are unable to determine when to intervene in a discussion, leading to undesirable frequent interventions and, consequently, derailment patterns similar to those observed in human interactions. Additionally, we find that different facilitation strategies influence conversational dynamics to some extent. Beyond our theoretical SDG framework, we also present a cost-comparison methodology for experimental design, an exploration of available models and algorithms, an open-source Python framework, and a large, publicly available dataset of LLM-generated discussions across multiple models.

cs.HC↗

arXiv2025

Evaluation and Facilitation of Online Discussions in the LLM Era: A Survey

We present a survey of methods for assessing and enhancing the quality of online discussions, focusing on the potential of LLMs. While online discourses aim, at least in theory, to foster mutual understanding, they often devolve into harmful exchanges, such as hate speech, threatening social cohesion and democratic values. Recent advancements in LLMs enable artificial facilitation agents to not only moderate content, but also actively improve the quality of interactions. Our survey synthesizes ideas from NLP and Social Sciences to provide (a) a new taxonomy on discussion quality evaluation, (b) an overview of intervention and facilitation strategies, (c) along with a new taxonomy of conversation facilitation datasets, (d) an LLM-oriented roadmap of good practices and future research directions, from technological and societal perspectives.

cs.CL↗

arXiv2025

Open or Closed LLM for Lesser-Resourced Languages? Lessons from Greek

Natural Language Processing (NLP) for lesser-resourced languages faces persistent challenges, including limited datasets, inherited biases from high-resource languages, and the need for domain-specific solutions. This study addresses these gaps for Modern Greek through three key contributions. First, we evaluate the performance of open-source (Llama-70b) and closed-source (GPT-4o mini) large language models (LLMs) on seven core NLP tasks with dataset availability, revealing task-specific strengths, weaknesses, and parity in their performance. Second, we expand the scope of Greek NLP by reframing Authorship Attribution as a tool to assess potential data usage by LLMs in pre-training, with high 0-shot accuracy suggesting ethical implications for data provenance. Third, we showcase a legal NLP case study, where a Summarize, Translate, and Embed (STE) methodology outperforms the traditional TF-IDF approach for clustering \emph{long} legal texts. Together, these contributions provide a roadmap to advance NLP in lesser-resourced languages, bridging gaps in model evaluation, task innovation, and real-world impact.

cs.CL↗

arXiv2024

GR-NLP-TOOLKIT: An Open-Source NLP Toolkit for Modern Greek

We present GR-NLP-TOOLKIT, an open-source natural language processing (NLP) toolkit developed specifically for modern Greek. The toolkit provides state-of-the-art performance in five core NLP tasks, namely part-of-speech tagging, morphological tagging, dependency parsing, named entity recognition, and Greeklishto-Greek transliteration. The toolkit is based on pre-trained Transformers, it is freely available, and can be easily installed in Python (pip install gr-nlp-toolkit). It is also accessible through a demonstration platform on HuggingFace, along with a publicly available API for non-commercial use. We discuss the functionality provided for each task, the underlying methods, experiments against comparable open-source toolkits, and future possible enhancements. The toolkit is available at: https://github.com/nlpaueb/gr-nlp-toolkit

cs.CL↗

arXiv2024

Hate Speech According to the Law: An Analysis for Effective Detection

The issue of hate speech extends beyond the confines of the online realm. It is a problem with real-life repercussions, prompting most nations to formulate legal frameworks that classify hate speech as a punishable offence. These legal frameworks differ from one country to another, contributing to the big chaos that online platforms have to face when addressing reported instances of hate speech. With the definitions of hate speech falling short in introducing a robust framework, we turn our gaze onto hate speech laws. We consult the opinion of legal experts on a hate speech dataset and we experiment by employing various approaches such as pretrained models both on hate speech and legal data, as well as exploiting two large language models (Qwen2-7B-Instruct and Meta-Llama-3-70B). Due to the time-consuming nature of data acquisition for prosecutable hate speech, we use pseudo-labeling to improve our pretrained models. This study highlights the importance of amplifying research on prosecutable hate speech and provides insights into effective strategies for combating hate speech within the parameters of legal frameworks. Our findings show that legal knowledge in the form of annotations can be useful when classifying prosecutable hate speech, yet more focus should be paid on the differences between the laws.

cs.CL↗