Hasil untuk "Greek philology and language"

Yang Zhang, Mersin Konomi, Christos Xypolopoulos et al.

Large Language Models (LLMs) are commonly trained on multilingual corpora that include Greek, yet reliable evaluation benchmarks for Greek-particularly those based on authentic, native-sourced content-remain limited. Existing datasets are often machine-translated from English, failing to capture Greek linguistic and cultural characteristics. We introduce GreekMMLU, a native-sourced benchmark for massive multitask language understanding in Greek, comprising 21,805 multiple-choice questions across 45 subject areas, organized under a newly defined subject taxonomy and annotated with educational difficulty levels spanning primary to professional examinations. All questions are sourced or authored in Greek from academic, professional, and governmental exams. We publicly release 16,857 samples and reserve 4,948 samples for a private leaderboard to enable robust and contamination-resistant evaluation. Evaluations of over 80 open- and closed-source LLMs reveal substantial performance gaps between frontier and open-weight models, as well as between Greek-adapted models and general multilingual ones. Finally, we provide a systematic analysis of factors influencing performance-including model scale, adaptation, and prompting-and derive insights for improving LLM capabilities in Greek.

John Pavlopoulos, Juli Bakagianni, Kanella Pouli et al.

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.

Nicolas Zucchet, Jörg Bornschein, Stephanie Chan et al.

Large language models accumulate vast knowledge during pre-training, yet the dynamics governing this acquisition remain poorly understood. This work investigates the learning dynamics of language models on a synthetic factual recall task, uncovering three key findings: First, language models learn in three phases, exhibiting a performance plateau before acquiring precise factual knowledge. Mechanistically, this plateau coincides with the formation of attention-based circuits that support recall. Second, the training data distribution significantly impacts learning dynamics, as imbalanced distributions lead to shorter plateaus. Finally, hallucinations emerge simultaneously with knowledge, and integrating new knowledge into the model through fine-tuning is challenging, as it quickly corrupts its existing parametric memories. Our results emphasize the importance of data distribution in knowledge acquisition and suggest novel data scheduling strategies to accelerate neural network training.

Dimitris Roussis, Leon Voukoutis, Georgios Paraskevopoulos et al.

We introduce Llama-Krikri-8B, a cutting-edge Large Language Model tailored for the Greek language, built on Meta's Llama 3.1-8B. Llama-Krikri-8B has been extensively trained on high-quality Greek data to ensure superior adaptation to linguistic nuances. With 8 billion parameters, it offers advanced capabilities while maintaining efficient computational performance. Llama-Krikri-8B supports both Modern Greek and English, and is also equipped to handle polytonic text and Ancient Greek. The chat version of Llama-Krikri-8B features a multi-stage post-training pipeline, utilizing both human and synthetic instruction and preference data, by applying techniques such as MAGPIE. In addition, for evaluation, we propose three novel public benchmarks for Greek. Our evaluation on existing as well as the proposed benchmarks shows notable improvements over comparable Greek and multilingual LLMs in both natural language understanding and generation as well as code generation.

Xueqing Peng, Triantafillos Papadopoulos, Efstathia Soufleri et al.

Despite Greece's pivotal role in the global economy, large language models (LLMs) remain underexplored for Greek financial context due to the linguistic complexity of Greek and the scarcity of domain-specific datasets. Previous efforts in multilingual financial natural language processing (NLP) have exposed considerable performance disparities, yet no dedicated Greek financial benchmarks or Greek-specific financial LLMs have been developed until now. To bridge this gap, we introduce Plutus-ben, the first Greek Financial Evaluation Benchmark, and Plutus-8B, the pioneering Greek Financial LLM, fine-tuned with Greek domain-specific data. Plutus-ben addresses five core financial NLP tasks in Greek: numeric and textual named entity recognition, question answering, abstractive summarization, and topic classification, thereby facilitating systematic and reproducible LLM assessments. To underpin these tasks, we present three novel, high-quality Greek financial datasets, thoroughly annotated by expert native Greek speakers, augmented by two existing resources. Our comprehensive evaluation of 22 LLMs on Plutus-ben reveals that Greek financial NLP remains challenging due to linguistic complexity, domain-specific terminology, and financial reasoning gaps. These findings underscore the limitations of cross-lingual transfer, the necessity for financial expertise in Greek-trained models, and the challenges of adapting financial LLMs to Greek text. We release Plutus-ben, Plutus-8B, and all associated datasets publicly to promote reproducible research and advance Greek financial NLP, fostering broader multilingual inclusivity in finance.

A. Lianeri

This article argues that classical philology can play a vital role in debates about the importance of philology now and configures a genealogy that may contribute to the quest for alternative philologies. Building on Werner Hamacher's definition of philology as “love of the non sequitur ,” I turn to founding texts of Western classical philology by Johann Joachim Winckelmann, Friedrich August Wolf, and August Böckh in order to interrogate their identification with modern classicism and historicism. Examining the science of philology as Altertumswissenschaft , I focus on a language of ambiguity and undecidability with regard to philology's classical object (Greek and Roman pasts) and the discourse of philological science that constructs it. This is grasped as the relation between a transcendental temporality that enunciated classical antiquity's wholeness and a kind of perturbation of time that destabilized philology's alignment with classicism and historicism. For Winckelmann, Wolf, and Böckh, the philologist's task required a conceptual and temporal leap toward the past that signaled the absence of philology's grounding. In this sense, it differed from evocations of a seamless movement across a unified horizon of time linking antiquity and modernity. This was conveyed by stressing the past's mutilation, absence, and accidental expression as the vanishing ground on which philology could build its classical vision. By configuring these notions as the self‐hollowing basis of its knowledge, classical philology came to be divided by a paradoxical appeal to sequential time and the non sequitur. Tensions produced in this context bring classical philology to the center of debates that seek to interrogate modern historical intelligibility and time. Far from perpetuating ideas of irreversible and linear time, classical philology claims to engage with an absent past, and as such, it disrupts sequential temporalities by desiring something that is always beyond presence or reach and therefore always available for times to come and for future emancipation from regimes of present time.

Luciana Boban, Josip Grubeša, Jelena Jurčić

Considering the importance of higher education for the development of all scientific disciplines, especially in the context of the higher education reform known as the “Bologna Reform”, it is surprising that there are no studies addressing classical philology in relation to higher education, and that this topic is generally not of interest to classical philologists. This paper analyzes the extent to which the fundamental elements of the reform influence the development of competencies in classical philologists, and thus classical philology as a whole, through two indicators: the level of regulation of the profession (using as an example the document Subject Benchmark Statement: Classics and Ancient History (including Byzantine Studies and Modern Greek)), and the method of monitoring the learning outcomes achievement (using as an example Latin Language and Roman Literature - undergraduate double-major study programme at the Faculty of Humanities and Social Sciences, University of Mostar). The core documents of the Bologna Reform are the European Qualifications Framework (EQF), whose elements serve as prerequisites for regulating professions, and the Standards and Guidelines for Quality Assurance in the European Higher Education Area (ESG), which, among other things, define quality standards for higher education institutions and their study programmes across the entire European Higher Education Area (EHEA). Since the emphasis is placed on the competencies of future professionals – classical philologists, i.e., current students of classical philology – this analysis draws from the ESG only those standards that are related to student assessment and the monitoring of the defined learning outcomes achievement. Keywords: competencies of classical philologists, Bologna Reform, assessment standards, learning outcomes, Subject Benchmark Statement, University of Mostar

Lefteris Loukas, Nikolaos Smyrnioudis, Chrysa Dikonomaki et al.

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

Katerina Papantoniou, Yannis Tzitzikas

English language is in the spotlight of the Natural Language Processing (NLP) community with other languages, like Greek, lagging behind in terms of offered methods, tools and resources. Due to the increasing interest in NLP, in this paper we try to condense research efforts for the automatic processing of Greek language covering the last three decades. In particular, we list and briefly discuss related works, resources and tools, categorized according to various processing layers and contexts. We are not restricted to the modern form of Greek language but also cover Ancient Greek and various Greek dialects. This survey can be useful for researchers and students interested in NLP tasks, Information Retrieval and Knowledge Management for the Greek language.

Leon Voukoutis, Dimitris Roussis, Georgios Paraskevopoulos et al.

We describe the development and capabilities of Meltemi 7B, the first open Large Language Model for the Greek language. Meltemi 7B has 7 billion parameters and is trained on a 40 billion token Greek corpus. For the development of Meltemi 7B, we adapt Mistral, by continuous pretraining on the Greek Corpus. Meltemi 7B contains up-to-date information up to September 2023. Furthermore, we have translated and curated a Greek instruction corpus, which has been used for the instruction-tuning of a chat model, named Meltemi 7B Instruct. Special care has been given to the alignment and the removal of toxic content for the Meltemi 7B Instruct. The developed models are evaluated on a broad set of collected evaluation corpora, and examples of prompts and responses are presented. Both Meltemi 7B and Meltemi 7B Instruct are available at https://huggingface.co/ilsp under the Apache 2.0 license.

Eric Zhang, Leshem Chosen, Jacob Andreas

When language models (LMs) are trained to forget (or "unlearn'') a skill, how precisely does their behavior change? We study the behavior of transformer LMs in which tasks have been forgotten via fine-tuning on randomized labels. Such LMs learn to generate near-random predictions for individual examples in the "training'' set used for forgetting. Across tasks, however, LMs exhibit extreme variability in whether LM predictions change on examples outside the training set. In some tasks (like entailment classification), forgetting generalizes robustly, and causes models to produce uninformative predictions on new task instances; in other tasks (like physical commonsense reasoning and scientific question answering) forgetting affects only the training examples, and models continue to perform the "forgotten'' task accurately even for examples very similar to those that appeared in the training set. Dataset difficulty is not predictive of whether a behavior can be forgotten; instead, generalization in forgetting is (weakly) predicted by the confidence of LMs' initial task predictions and the variability of LM representations of training data, with low confidence and low variability both associated with greater generalization. Perhaps most surprisingly, random-label forgetting appears to be somewhat insensitive to the contents of the training set: for example, models trained on science questions with random labels continue to answer other science questions accurately, but begin to produce random labels on entailment classification tasks. Finally, we show that even generalizable forgetting is shallow: linear probes trained on LMs' representations can still perform tasks reliably after forgetting. Our results highlight the difficulty and unpredictability of performing targeted skill removal from models via fine-tuning.

Juli Bakagianni, Kanella Pouli, Maria Gavriilidou et al.

Comprehensive monolingual Natural Language Processing (NLP) surveys are essential for assessing language-specific challenges, resource availability, and research gaps. However, existing surveys often lack standardized methodologies, leading to selection bias and fragmented coverage of NLP tasks and resources. This study introduces a generalizable framework for systematic monolingual NLP surveys. Our approach integrates a structured search protocol to minimize bias, an NLP task taxonomy for classification, and language resource taxonomies to identify potential benchmarks and highlight opportunities for improving resource availability. We apply this framework to Greek NLP (2012-2023), providing an in-depth analysis of its current state, task-specific progress, and resource gaps. The survey results are publicly available (https://doi.org/10.5281/zenodo.15314882) and are regularly updated to provide an evergreen resource. This systematic survey of Greek NLP serves as a case study, demonstrating the effectiveness of our framework and its potential for broader application to other not so well-resourced languages as regards NLP.

M. P. Parkhaeva

Maja Baćić-Ćosić, Anka Rađenović

Prepositions are indeclinable words with limited lexical meaning that cannot stand alone but can govern one or more cases. In Modern Greek, which has four cases (nominative, genitive, accusative, and vocative), prepositions are commonly used to express a variety of relations (such as location, time, direction, etc.). Specifically, certain prepositions in this language can be followed simultaneously by the accusative and genitive cases. The aim of this paper is to investigate how a group of students of Modern Greek as L2 at the Department of Modern Greek Studies, Faculty of Philology, University of Belgrade, perceive the use of prepositions that syntactically correspond to the accusative and the genitive and change their meaning depending on the case they are used with. A non-experimental quantitative survey with multiple-choice, closed-ended questions was conducted. Respondents were asked to form prepositional phrases with prepositions that can be followed by both genitive and accusative (epί, ypό, apό, pros, metά, and catά) by choosing nouns in one of the above cases. This paper aims to identify the semantic and syntactic components that may be problematic for learners of Modern Greek as L2 in the use of prepositions and prepositional phrases, as well as to suggest strategies for more efficient acquisition and use of this word class in Modern Greek.

Gonzalo Jerez Sánchez, Unai Iriarte

The study of ancient Greek has faced significant challenges, from its perceived lack of practicality to the apparent obsolescence of Classical Philology. The teaching of Greek has evolved due to historical and political developments in countries where it was taught. Currently, the value of ancient Greek in educational curricula is under debate, threatening its future. Despite its limited contemporary applications, ancient Greek remains essential for scholars of Classical Philology, history, and Eastern European studies. We propose exploring alternative methods to approach the language, such as those by Moschopulos, utilizing texts of various categories and purposes, and reflecting on historical teaching methods. By moving away from current positivist stances, we can revitalize the study of this language and offer new perspectives.

Frederick Riemenschneider, A. Frank

Recent advances in NLP have led to the creation of powerful language models for many languages including Ancient Greek and Latin. While prior work on Classical languages unanimously uses BERT, in this work we create four language models for Ancient Greek that vary along two dimensions to study their versatility for tasks of interest for Classical languages: we explore (i) encoder-only and encoder-decoder architectures using RoBERTa and T5 as strong model types, and create for each of them (ii) a monolingual Ancient Greek and a multilingual instance that includes Latin and English. We evaluate all models on morphological and syntactic tasks, including lemmatization, which demonstrates the added value of T5’s decoding abilities. We further define two probing tasks to investigate the knowledge acquired by models pre-trained on Classical texts. Our experiments provide the first benchmarking analysis of existing models of Ancient Greek. Results show that our models provide significant improvements over the SoTA. The systematic analysis of model types can inform future research in designing language models for Classical languages, including the development of novel generative tasks. We make all our models available as community resources, along with a large curated pre-training corpus for Ancient Greek, to support the creation of a larger, comparable model zoo for Classical Philology.

Tariq A. Yousef, Chiara Palladino, Farnoosh Shamsian

Jochen Sauer

Susanne Aretz

María Emilia Cairo

Revisión del libro Eneida. Virgilio por Estefanía, D.