Hasil untuk "Greek philology and language"

Journal Neograeca Bohemica

Susanne Aretz

Petros Marazopoulos

In my paper, I discuss the Greek literary responses to the phenomenon of the economic crisis that erupted in 2008. Dystopian narratives appear to be a popular narrative framework for authors dealing with the economic crisis during the decade of 2010. The purpose of this article is to examine contemporary Greek literary texts that represent societies under economic crisis and were written after the recent recession. Greek authors often used dystopian narratives to depict the relationship between economic politics followed by the European Union, the world system and authoritarianism. These literary works reflect the authors' experiences of the concurrent political and economic reality, serving both as literary responses to the economic system as well as critique to such systems. In addition, the literary texts under examination raise concerns regarding the growing power of economic institutions and their interaction with political power. Greek authors oft en construct claustrophobic narrative structures, to explore themes such as the erosion of individuality, pervasive state surveillance and escalating oppression. In that sense, I explore literary representations of "Economy", "Power", "Europe" and "Work" in times of austerity, discussing them in the wider context of dystopian fiction and the theory of totalitarianism.

Uli Sauerland, Celia Matthaei, Felix Salfner

We argue that human language learning proceeds in a manner that is different in nature from current approaches to training LLMs, predicting a difference in learning biases. We then present evidence from German plural formation by LLMs that confirm our hypothesis that even very powerful implementations produce results that miss aspects of the logic inherent to language that humans have no problem with. We conclude that attention to the different structures of human language and artificial neural networks is likely to be an avenue to improve LLM performance.

Jeffrey S. Bowers, Jeff Mitchell

According to Futrell and Mahowald [arXiv:2501.17047], both infants and language models (LMs) find attested languages easier to learn than impossible languages that have unnatural structures. We review the literature and show that LMs often learn attested and many impossible languages equally well. Difficult to learn impossible languages are simply more complex (or random). LMs are missing human inductive biases that support language acquisition.

Friedrich Maier

Susanne Aretz

Niyati Bafna, Kenton Murray, David Yarowsky

While large language models exhibit certain cross-lingual generalization capabilities, they suffer from performance degradation (PD) on unseen closely-related languages (CRLs) and dialects relative to their high-resource language neighbour (HRLN). However, we currently lack a fundamental understanding of what kinds of linguistic distances contribute to PD, and to what extent. Furthermore, studies of cross-lingual generalization are confounded by unknown quantities of CRL language traces in the training data, and by the frequent lack of availability of evaluation data in lower-resource related languages and dialects. To address these issues, we model phonological, morphological, and lexical distance as Bayesian noise processes to synthesize artificial languages that are controllably distant from the HRLN. We analyse PD as a function of underlying noise parameters, offering insights on model robustness to isolated and composed linguistic phenomena, and the impact of task and HRL characteristics on PD. We calculate parameter posteriors on real CRL-HRLN pair data and show that they follow computed trends of artificial languages, demonstrating the viability of our noisers. Our framework offers a cheap solution for estimating task performance on an unseen CRL given HRLN performance using its posteriors, as well as for diagnosing observed PD on a CRL in terms of its linguistic distances from its HRLN, and opens doors to principled methods of mitigating performance degradation.

Xiaonan Wang, Jinyoung Yeo, Joon-Ho Lim et al.

Large language models have exhibited significant enhancements in performance across various tasks. However, the complexity of their evaluation increases as these models generate more fluent and coherent content. Current multilingual benchmarks often use translated English versions, which may incorporate Western cultural biases that do not accurately assess other languages and cultures. To address this research gap, we introduce KULTURE Bench, an evaluation framework specifically designed for Korean culture that features datasets of cultural news, idioms, and poetry. It is designed to assess language models' cultural comprehension and reasoning capabilities at the word, sentence, and paragraph levels. Using the KULTURE Bench, we assessed the capabilities of models trained with different language corpora and analyzed the results comprehensively. The results show that there is still significant room for improvement in the models' understanding of texts related to the deeper aspects of Korean culture.

Ziqiao Ma, Zekun Wang, Joyce Chai

Humans are efficient language learners and inherently social creatures. Our language development is largely shaped by our social interactions, for example, the demonstration and feedback from caregivers. Contrary to human language learning, recent advancements in large language models have primarily adopted a non-interactive training paradigm, and refined pre-trained models through feedback afterward. In this work, we explore how corrective feedback from interactions influences neural language acquisition from scratch through systematically controlled experiments, assessing whether it contributes to word learning efficiency in language models. We introduce a trial-and-demonstration (TnD) learning framework that incorporates three distinct components: student trials, teacher demonstrations, and a reward conditioned on language competence at various developmental stages. Our experiments reveal that the TnD approach accelerates word acquisition for student models of equal and smaller numbers of parameters, and we highlight the significance of both trials and demonstrations. We further show that the teacher's choices of words influence students' word-specific learning efficiency, and a practice-makes-perfect effect is evident by a strong correlation between the frequency of words in trials and their respective learning curves. Our findings suggest that interactive language learning, with teacher demonstrations and active trials, can facilitate efficient word learning in language models.

Joachim Penzel

Wenxuan Wang, Zhaopeng Tu, Chang Chen et al.

Safety lies at the core of developing and deploying large language models (LLMs). However, previous safety benchmarks only concern the safety in one language, e.g. the majority language in the pretraining data such as English. In this work, we build the first multilingual safety benchmark for LLMs, XSafety, in response to the global deployment of LLMs in practice. XSafety covers 14 kinds of commonly used safety issues across 10 languages that span several language families. We utilize XSafety to empirically study the multilingual safety for 4 widely-used LLMs, including both close-API and open-source models. Experimental results show that all LLMs produce significantly more unsafe responses for non-English queries than English ones, indicating the necessity of developing safety alignment for non-English languages. In addition, we propose several simple and effective prompting methods to improve the multilingual safety of ChatGPT by evoking safety knowledge and improving cross-lingual generalization of safety alignment. Our prompting method can significantly reduce the ratio of unsafe responses from 19.1% to 9.7% for non-English queries. We release our data at https://github.com/Jarviswang94/Multilingual_safety_benchmark.

Sarah Brauckmann

Zuzana Dzurillová

The present article explores the usage of the historical present tense in the narration of the Early Modern Greek romance Erotokritos. Focusing on both the shape and the semantics of present indicatives designating past events in a narrative context, the analysis investigates the phenomenon from the perspective of diegetic and mimetic narrative modes and the respective discourse-pragmatic functions. First, the examination demonstrates the diegetic quality of the historical present tense in summary narratives to foreshadow cognitively salient events in the story and focalize referents important to the plot. Second, it elucidates the tense's mimetic ability in scenic narratives to create a dramatic atmosphere, and third, it illuminates the static dimension of this technique with the tendency to express mental states and psychological expressions of the protagonists. The analysis provides evidence of the use of the historical present tense as a cultivated literary device from both narratological and philological points of view, shedding new light on the diachronic development of this phenomenon.

Stefano Costa, SC, Chiara Schürch, CS, Fabio Bellorio, FB et al.

CICERONE, In difesa di Archia, saggio introduttivo, nuova traduzione e note a cura di Daniele PELLACANI (S. COSTA)                 143 Gernot Michael MÜLLER, Jörn MÜLLER (Hgg.), Cicero ethicus. Die Tusculanae disputationes im Vergleich mit De finibus bonorum et malorum (C. Schürch)           147 Christopher DIEZ, Ciceros emanzipatorische Leserführung. Studien zum Verhältnis von dialogisch-rhetorischer Inszenierung und skeptischer Philosophie in De natura deorum (F. BELLORIO)    153 Tommaso GAZZARRI, The Stylus and the Scalpel. Theory and Pratice of Metaphors in Seneca’s Prose (A. CASAMENTO)  159

Ahmet Yavuz Uluslu, Gerold Schneider

Native language identification (NLI) is the task of automatically identifying the native language (L1) of an individual based on their language production in a learned language. It is useful for a variety of purposes including marketing, security and educational applications. NLI is usually framed as a multi-label classification task, where numerous designed features are combined to achieve state-of-the-art results. Recently deep generative approach based on transformer decoders (GPT-2) outperformed its counterparts and achieved the best results on the NLI benchmark datasets. We investigate this approach to determine the practical implications compared to traditional state-of-the-art NLI systems. We introduce transformer adapters to address memory limitations and improve training/inference speed to scale NLI applications for production.

Konstantina Dritsa, Kaiti Thoma, John Pavlopoulos et al.

Large, diachronic datasets of political discourse are hard to come across, especially for resource-lean languages such as Greek. In this paper, we introduce a curated dataset of the Greek Parliament Proceedings that extends chronologically from 1989 up to 2020. It consists of more than 1 million speeches with extensive metadata, extracted from 5,355 parliamentary record files. We explain how it was constructed and the challenges that we had to overcome. The dataset can be used for both computational linguistics and political analysis-ideally, combining the two. We present such an application, showing (i) how the dataset can be used to study the change of word usage through time, (ii) between significant historical events and political parties, (iii) by evaluating and employing algorithms for detecting semantic shifts.

Jesin James, Isabella Shields, Vithya Yogarajan et al.

Te reo Māori (referred to as Māori), New Zealand's indigenous language, is under-resourced in language technology. Māori speakers are bilingual, where Māori is code-switched with English. Unfortunately, there are minimal resources available for Māori language technology, language detection and code-switch detection between Māori-English pair. Both English and Māori use Roman-derived orthography making rule-based systems for detecting language and code-switching restrictive. Most Māori language detection is done manually by language experts. This research builds a Māori-English bilingual database of 66,016,807 words with word-level language annotation. The New Zealand Parliament Hansard debates reports were used to build the database. The language labels are assigned using language-specific rules and expert manual annotations. Words with the same spelling, but different meanings, exist for Māori and English. These words could not be categorised as Māori or English based on word-level language rules. Hence, manual annotations were necessary. An analysis reporting the various aspects of the database such as metadata, year-wise analysis, frequently occurring words, sentence length and N-grams is also reported. The database developed here is a valuable tool for future language and speech technology development for Aotearoa New Zealand. The methodology followed to label the database can also be followed by other low-resourced language pairs.

Wenlong Huang, Pieter Abbeel, Deepak Pathak et al.

Can world knowledge learned by large language models (LLMs) be used to act in interactive environments? In this paper, we investigate the possibility of grounding high-level tasks, expressed in natural language (e.g. "make breakfast"), to a chosen set of actionable steps (e.g. "open fridge"). While prior work focused on learning from explicit step-by-step examples of how to act, we surprisingly find that if pre-trained LMs are large enough and prompted appropriately, they can effectively decompose high-level tasks into mid-level plans without any further training. However, the plans produced naively by LLMs often cannot map precisely to admissible actions. We propose a procedure that conditions on existing demonstrations and semantically translates the plans to admissible actions. Our evaluation in the recent VirtualHome environment shows that the resulting method substantially improves executability over the LLM baseline. The conducted human evaluation reveals a trade-off between executability and correctness but shows a promising sign towards extracting actionable knowledge from language models. Website at https://huangwl18.github.io/language-planner

Fatemeh Azadi, Heshaam Faili, Mohammad Javad Dousti

Translation Quality Estimation (QE) is the task of predicting the quality of machine translation (MT) output without any reference. This task has gained increasing attention as an important component in the practical applications of MT. In this paper, we first propose XLMRScore, which is a cross-lingual counterpart of BERTScore computed via the XLM-RoBERTa (XLMR) model. This metric can be used as a simple unsupervised QE method, nevertheless facing two issues: firstly, the untranslated tokens leading to unexpectedly high translation scores, and secondly, the issue of mismatching errors between source and hypothesis tokens when applying the greedy matching in XLMRScore. To mitigate these issues, we suggest replacing untranslated words with the unknown token and the cross-lingual alignment of the pre-trained model to represent aligned words closer to each other, respectively. We evaluate the proposed method on four low-resource language pairs of the WMT21 QE shared task, as well as a new English$\rightarrow$Persian (En-Fa) test dataset introduced in this paper. Experiments show that our method could get comparable results with the supervised baseline for two zero-shot scenarios, i.e., with less than 0.01 difference in Pearson correlation, while outperforming unsupervised rivals in all the low-resource language pairs for above 8%, on average.