Hasil untuk "Greek philology and language"

Yasmin Moslem, Aman Kassahun Wassie, Amanuel Gizachew Abebe

In this work, we present AfriNLLB, a series of lightweight models for efficient translation from and into African languages. AfriNLLB supports 15 language pairs (30 translation directions), including Swahili, Hausa, Yoruba, Amharic, Somali, Zulu, Lingala, Afrikaans, Wolof, and Egyptian Arabic, as well as other African Union official languages such as Arabic (MSA), French, Portuguese, and Spanish. Our training data covers bidirectional translation between English and 13 languages, and between French and two languages (Lingala and Wolof). AfriNLLB models are based on NLLB-200 600M, which we compress using iterative layer pruning and quantization. We fine-tune the pruned models on parallel corpora we curated for African languages, employing knowledge distillation from a larger teacher model. Our work aims at enabling efficient deployment of translation models for African languages in resource-constrained settings. Our evaluation results demonstrate that AfriNLLB models achieve performance comparable to the baseline while being significantly faster. We release two versions of the AfriNLLB models, a Transformers version that allows further fine-tuning and a CTranslate2 version for efficient inference. Moreover, we release all the training data that we used for fine-tuning the baseline and pruned models to facilitate further research.

Pedazzini, Han

La stele riporta un decreto ateniese del 357/356 che predispone misure economiche e militari per Andro, alleata di Atene: il finanziamento di una guarnigione già presente sull’isola tramite le syntaxeis degli alleati; l’invio di uno stratego; l’incarico di raccogliere il denaro dalle isole vicine. Questo rappresenterebbe una dichiarata infrazione dello statuto della Seconda lega ateniese, che proibiva il ritorno alle pratiche della Lega delio-attica. Il documento, nel contesto Guerra Sociale, deve essere considerato, invece, come la comunicazione di una decisione richiesta e condivisa con gli Alleati volta alla sicurezza di Andro.

Dominic Akt, Marco Peer, Florian Kleber

This paper tackles the task of writer identification for Greek papyri. A common preprocessing step in writer identification pipelines is image binarization, which prevents the model from learning background features. This is challenging in historical documents, in our case Greek papyri, as background is often non-uniform, fragmented, and discolored with visible fiber structures. We compare traditional binarization methods to state-of-the-art Deep Learning (DL) models, evaluating the impact of binarization quality on subsequent writer identification performance. DL models are trained with and without a custom data augmentation technique, as well as different model selection criteria are applied. The performance of these binarization methods, is then systematically evaluated on the DIBCO 2019 dataset. The impact of binarization on writer identification is subsequently evaluated using a state-of-the-art approach for writer identification. The results of this analysis highlight the influence of data augmentation for DL methods. Furthermore, findings indicate a strong correlation between binarization effectiveness on papyri documents of DIBCO 2019 and downstream writer identification performance.

Kilian Sennrich, Sina Ahmadi

Knowledge graphs offer an excellent solution for representing the lexical-semantic structures of lexicographic data. However, working with the SPARQL query language represents a considerable hurdle for many non-expert users who could benefit from the advantages of this technology. This paper addresses the challenge of creating natural language interfaces for lexicographic data retrieval on knowledge graphs such as Wikidata. We develop a multidimensional taxonomy capturing the complexity of Wikidata's lexicographic data ontology module through four dimensions and create a template-based dataset with over 1.2 million mappings from natural language utterances to SPARQL queries. Our experiments with GPT-2 (124M), Phi-1.5 (1.3B), and GPT-3.5-Turbo reveal significant differences in model capabilities. While all models perform well on familiar patterns, only GPT-3.5-Turbo demonstrates meaningful generalization capabilities, suggesting that model size and diverse pre-training are crucial for adaptability in this domain. However, significant challenges remain in achieving robust generalization, handling diverse linguistic data, and developing scalable solutions that can accommodate the full complexity of lexicographic knowledge representation.

Daniel Christoph, Max Ploner, Patrick Haller et al.

Sample efficiency is a crucial property of language models with practical implications for training efficiency. In real-world text, information follows a long-tailed distribution. Yet, we expect models to learn and recall frequent and infrequent facts. Sample-efficient models are better equipped to handle this challenge of learning and retaining rare information without requiring excessive exposure. This study analyzes multiple models of varying architectures and sizes, all trained on the same pre-training data. By annotating relational facts with their frequencies in the training corpus, we examine how model performance varies with fact frequency. Our findings show that most models perform similarly on high-frequency facts but differ notably on low-frequency facts. This analysis provides new insights into the relationship between model architecture, size, and factual learning efficiency.

Susanne Aretz

Leonardo Ranaldi, Andrè Freitas

The alignments of reasoning abilities between smaller and larger Language Models are largely conducted via Supervised Fine-Tuning (SFT) using demonstrations generated from robust Large Language Models (LLMs). Although these approaches deliver more performant models, they do not show sufficiently strong generalization ability as the training only relies on the provided demonstrations. In this paper, we propose the Self-refine Instruction-tuning method that elicits Smaller Language Models to self-refine their abilities. Our approach is based on a two-stage process, where reasoning abilities are first transferred between LLMs and Small Language Models (SLMs) via Instruction-tuning on demonstrations provided by LLMs, and then the instructed models Self-refine their abilities through preference optimization strategies. In particular, the second phase operates refinement heuristics based on the Direct Preference Optimization algorithm, where the SLMs are elicited to deliver a series of reasoning paths by automatically sampling the generated responses and providing rewards using ground truths from the LLMs. Results obtained on commonsense and math reasoning tasks show that this approach significantly outperforms Instruction-tuning in both in-domain and out-domain scenarios, aligning the reasoning abilities of Smaller and Larger Language Models.

Andre Rusli, Makoto Shishido

This research explores the applicability of cross-lingual transfer learning from English to Japanese and Indonesian using the XLM-R pre-trained model. The results are compared with several previous works, either by models using a similar zero-shot approach or a fully-supervised approach, to provide an overview of the zero-shot transfer learning approach's capability using XLM-R in comparison with existing models. Our models achieve the best result in one Japanese dataset and comparable results in other datasets in Japanese and Indonesian languages without being trained using the target language. Furthermore, the results suggest that it is possible to train a multi-lingual model, instead of one model for each language, and achieve promising results.

Marco Peer, Robert Sablatnig, Olga Serbaeva et al.

This paper presents a character-based approach for enhancing writer retrieval performance in the context of Greek papyri. Our contribution lies in introducing character-level annotations for frequently used characters, in our case the trigram kai and four additional letters (epsilon, kappa, mu, omega), in Greek texts. We use a state-of-the-art writer retrieval approach based on NetVLAD and compare a character-level-based feature aggregation method against the current default baseline of using small patches located at SIFT keypoint locations for building the page descriptors. We demonstrate that by using only about 15 characters per page, we are able to boost the performance up to 4% mAP (a relative improvement of 11%) on the GRK-120 dataset. Additionally, our qualitative analysis offers insights into the similarity scores of SIFT patches and specific characters. We publish the dataset with character-level annotations, including a quality label and our binarized images for further research.

Olga Iakovenko, Thomas Hain

Code-switching (CS) is the process of speakers interchanging between two or more languages which in the modern world becomes increasingly common. In order to better describe CS speech the Matrix Language Frame (MLF) theory introduces the concept of a Matrix Language, which is the language that provides the grammatical structure for a CS utterance. In this work the MLF theory was used to develop systems for Matrix Language Identity (MLID) determination. The MLID of English/Mandarin and English/Spanish CS text and speech was compared to acoustic language identity (LID), which is a typical way to identify a language in monolingual utterances. MLID predictors from audio show higher correlation with the textual principles than LID in all cases while also outperforming LID in an MLID recognition task based on F1 macro (60%) and correlation score (0.38). This novel approach has identified that non-English languages (Mandarin and Spanish) are preferred over the English language as the ML contrary to the monolingual choice of LID.

Oluwatosin Ogundare, Subuola Sofolahan

This study investigates the potential of an ambulatory device that incorporates Large Language Models (LLMs) in cadence with other specialized ML models to assess anemia severity in sickle cell patients in real time. The device would rely on sensor data that measures angiogenic material levels to assess anemia severity, providing real-time information to patients and clinicians to reduce the frequency of vaso-occlusive crises because of the early detection of anemia severity, allowing for timely interventions and potentially reducing the likelihood of serious complications. The main challenges in developing such a device are the creation of a reliable non-invasive tool for angiogenic level assessment, a biophysics model and the practical consideration of an LLM communicating with emergency personnel on behalf of an incapacitated patient. A possible system is proposed, and the limitations of this approach are discussed.

Ilias Dimitriadis, Dimitrios P. Giakatos, Stelios Karamanidis et al.

In this paper, we study the Greek wiretappings scandal, which has been revealed in 2022 and attracted a lot of attention by press and citizens. Specifically, we propose a methodology for collecting data and analyzing patterns of online public discussions on Twitter. We apply our methodology to the Greek wiretappings use case, and present findings related to the evolution of the discussion over time, its polarization, and the role of the media. The methodology can be of wider use and replicated to other topics. Finally, we provide publicly an open dataset, and online resources with the results.

Yaoyiran Li, Anna Korhonen, Ivan Vulić

Bilingual Lexicon Induction (BLI) is a core task in multilingual NLP that still, to a large extent, relies on calculating cross-lingual word representations. Inspired by the global paradigm shift in NLP towards Large Language Models (LLMs), we examine the potential of the latest generation of LLMs for the development of bilingual lexicons. We ask the following research question: Is it possible to prompt and fine-tune multilingual LLMs (mLLMs) for BLI, and how does this approach compare against and complement current BLI approaches? To this end, we systematically study 1) zero-shot prompting for unsupervised BLI and 2) few-shot in-context prompting with a set of seed translation pairs, both without any LLM fine-tuning, as well as 3) standard BLI-oriented fine-tuning of smaller LLMs. We experiment with 18 open-source text-to-text mLLMs of different sizes (from 0.3B to 13B parameters) on two standard BLI benchmarks covering a range of typologically diverse languages. Our work is the first to demonstrate strong BLI capabilities of text-to-text mLLMs. The results reveal that few-shot prompting with in-context examples from nearest neighbours achieves the best performance, establishing new state-of-the-art BLI scores for many language pairs. We also conduct a series of in-depth analyses and ablation studies, providing more insights on BLI with (m)LLMs, also along with their limitations.

Aaron J. Snoswell, Lucinda Nelson, Hao Xue et al.

Generic `toxicity' classifiers continue to be used for evaluating the potential for harm in natural language generation, despite mounting evidence of their shortcomings. We consider the challenge of measuring misogyny in natural language generation, and argue that generic `toxicity' classifiers are inadequate for this task. We use data from two well-characterised `Incel' communities on Reddit that differ primarily in their degrees of misogyny to construct a pair of training corpora which we use to fine-tune two language models. We show that an open source `toxicity' classifier is unable to distinguish meaningfully between generations from these models. We contrast this with a misogyny-specific lexicon recently proposed by feminist subject-matter experts, demonstrating that, despite the limitations of simple lexicon-based approaches, this shows promise as a benchmark to evaluate language models for misogyny, and that it is sensitive enough to reveal the known differences in these Reddit communities. Our preliminary findings highlight the limitations of a generic approach to evaluating harms, and further emphasise the need for careful benchmark design and selection in natural language evaluation.

Susanne Aretz

María Celeste Carrettoni

“The Legend of Thisbe” reescribe el mito de Tisbe y Píramo. Basado en Las metamorfosis, el relato de Chaucer introduce dos modificaciones principales: como otros en The Legend of Good Women, omite la metamorfosis. Más importante aun, el marco ficcional del Prólogo obliga a reinterpretar el mito. Ello se debe en parte a que The Legend of Good Women es una palinodia, y se inserta en la tradición que debate las virtudes de los sexos, tal como atestiguan el Prólogo y las amplificaciones del relato, que invitan a reflexionar –no sin cierta ironía—acerca de la conducta de la protagonista.

Susanne Aretz, Thomas Doepner, Oliver Klewer et al.

Barack W. Wanjawa, Lilian D. A. Wanzare, Florence Indede et al.

The need for Question Answering datasets in low resource languages is the motivation of this research, leading to the development of Kencorpus Swahili Question Answering Dataset, KenSwQuAD. This dataset is annotated from raw story texts of Swahili low resource language, which is a predominantly spoken in Eastern African and in other parts of the world. Question Answering (QA) datasets are important for machine comprehension of natural language for tasks such as internet search and dialog systems. Machine learning systems need training data such as the gold standard Question Answering set developed in this research. The research engaged annotators to formulate QA pairs from Swahili texts collected by the Kencorpus project, a Kenyan languages corpus. The project annotated 1,445 texts from the total 2,585 texts with at least 5 QA pairs each, resulting into a final dataset of 7,526 QA pairs. A quality assurance set of 12.5% of the annotated texts confirmed that the QA pairs were all correctly annotated. A proof of concept on applying the set to the QA task confirmed that the dataset can be usable for such tasks. KenSwQuAD has also contributed to resourcing of the Swahili language.

Alex Warstadt, Samuel R. Bowman

Rapid progress in machine learning for natural language processing has the potential to transform debates about how humans learn language. However, the learning environments and biases of current artificial learners and humans diverge in ways that weaken the impact of the evidence obtained from learning simulations. For example, today's most effective neural language models are trained on roughly one thousand times the amount of linguistic data available to a typical child. To increase the relevance of learnability results from computational models, we need to train model learners without significant advantages over humans. If an appropriate model successfully acquires some target linguistic knowledge, it can provide a proof of concept that the target is learnable in a hypothesized human learning scenario. Plausible model learners will enable us to carry out experimental manipulations to make causal inferences about variables in the learning environment, and to rigorously test poverty-of-the-stimulus-style claims arguing for innate linguistic knowledge in humans on the basis of speculations about learnability. Comparable experiments will never be possible with human subjects due to practical and ethical considerations, making model learners an indispensable resource. So far, attempts to deprive current models of unfair advantages obtain sub-human results for key grammatical behaviors such as acceptability judgments. But before we can justifiably conclude that language learning requires more prior domain-specific knowledge than current models possess, we must first explore non-linguistic inputs in the form of multimodal stimuli and multi-agent interaction as ways to make our learners more efficient at learning from limited linguistic input.

Muhammad Humayoun

This paper presents the system descriptions submitted at the FIRE Shared Task 2021 on Urdu's Abusive and Threatening Language Detection Task. This challenge aims at automatically identifying abusive and threatening tweets written in Urdu. Our submitted results were selected for the third recognition at the competition. This paper reports a non-exhaustive list of experiments that allowed us to reach the submitted results. Moreover, after the result declaration of the competition, we managed to attain even better results than the submitted results. Our models achieved 0.8318 F1 score on Task A (Abusive Language Detection for Urdu Tweets) and 0.4931 F1 score on Task B (Threatening Language Detection for Urdu Tweets). Results show that Support Vector Machines with stopwords removed, lemmatization applied, and features vector created by the combinations of word n-grams for n=1,2,3 produced the best results for Task A. For Task B, Support Vector Machines with stopwords removed, lemmatization not applied, feature vector created from a pre-trained Urdu Word2Vec (on word unigrams and bigrams), and making the dataset balanced using oversampling technique produced the best results. The code is made available for reproducibility.