Hasil untuk "German literature"

Ashish Vaswani, Noam Shazeer, Niki Parmar et al.

The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.

Marco Baroni, Georgiana Dinu, Germán Kruszewski

Context-predicting models (more commonly known as embeddings or neural language models) are the new kids on the distributional semantics block. Despite the buzz surrounding these models, the literature is still lacking a systematic comparison of the predictive models with classic, count-vector-based distributional semantic approaches. In this paper, we perform such an extensive evaluation, on a wide range of lexical semantics tasks and across many parameter settings. The results, to our own surprise, show that the buzz is fully justified, as the context-predicting models obtain a thorough and resounding victory against their count-based counterparts.

D. Blazer

Phuong Chi Diep, Martin D. Hartmann

The primary objective of this study is to develop a comprehensive theoretical framework for the action-oriented teaching perspective, grounded in the German approach, widely adopted in German-speaking countries. The article proposes a process that teachers can use to design and implement action-oriented teaching to enhance students’ application of academic content in realistic professional situations and social contexts. This study follows a qualitative approach, employing systematic literature review, grounded theory, and thematic analysis methods. The findings reveal that action-oriented teaching has a strong psychological foundation. When analyzed in relation to cognitive and skills taxonomies, and empirical data from Germany and Vietnam, this concept demonstrates its capacity to foster learners’ complex competencies—particularly their ability to apply theoretical knowledge in professional and social contexts. Action-oriented teaching can be implemented through various methods, such as project-based learning, experimentation, design tasks, fabrication tasks, debates, role-plays, and case studies. A key aspect of this concept is that learners actively engage in comprehensive actions to create material (tangible) or verbal learning products, which result from situation-oriented learning tasks. This framework helps address key educational challenges, including student engagement, competency development, and the integration of theory and practice. Its interdisciplinary and cross-cultural relevance makes it valuable for educators, policymakers, and researchers across vocational, general, and higher education contexts.

Jonathan Drechsel, Erisa Bytyqi, Steffen Herbold

Language models perform well on grammatical agreement, but it is unclear whether this reflects rule-based generalization or memorization. We study this question for German definite singular articles, whose forms depend on gender and case. Using GRADIEND, a gradient-based interpretability method, we learn parameter update directions for gender-case specific article transitions. We find that updates learned for a specific gender-case article transition frequently affect unrelated gender-case settings, with substantial overlap among the most affected neurons across settings. These results argue against a strictly rule-based encoding of German definite articles, indicating that models at least partly rely on memorized associations rather than abstract grammatical rules.

Ali Hamza Bashir, Muhammad Rehan Khalid, Kostadin Cvejoski et al.

Large language models (LLMs) often struggle in specialized domains such as legal reasoning due to limited expert knowledge, resulting in factually incorrect outputs or hallucinations. This paper presents an effective method for adapting advanced LLMs to German legal question answering through a novel synthetic data generation approach. In contrast to costly human-annotated resources or unreliable synthetic alternatives, our approach systematically produces high-quality, diverse, and legally accurate question-answer pairs directly from authoritative German statutes. Using rigorous automated filtering methods and parameter-efficient fine-tuning techniques, we demonstrate that LLMs adapted with our synthetic dataset significantly outperform their baseline counterparts on German legal question answering tasks. Our results highlight the feasibility of using carefully designed synthetic data as a robust alternative to manual annotation in high-stakes, knowledge-intensive domains.

Carla Riviello

Nel Old English Boethius, i riferimenti al mito sono spesso introdotti o accompagnati dalla precisazione che si tratta di storie non vere, ma necessarie a spiegare importanti verità. Questo atteggiamento, comune alla cristianità medievale, è reso esplicito in diverse occasioni. Nel rispetto della funzione proposta nell’originale, il materiale mitologico viene quindi rielaborato attraverso omissioni e aggiunte per adattarsi alla specifica ricezione che l’opera avrebbe avuto nell’Inghilterra dell’età alfrediana. Il confronto con le possibili fonti considerate dalla critica offre ipotesi interessanti, mentre l’analisi dei singoli passaggi sembra rivelare strategie traduttive particolari nel riposizionamento di episodi che hanno come protagoniste figure come Orfeo, Ercole o i giganti. L’obiettivo di questo articolo è indagare sulle strategie traduttive adottate per la rappresentazione del mito nella rielaborazione in volgare: oltre alle esigenze didattiche, essa sembra essere guidata dal fascino di questi antichi racconti, ambientati in mondi geograficamente e cronologicamente ignoti e la cui veridicità non dev’essere creduta, ma che l’autore presenta nel dettaglio, condividendo con il pubblico il piacere di storie straordinarie e sorprendenti.

Dinh Nam Pham, Eleftherios Avramidis

Sign Language Recognition (SLR) systems primarily focus on manual gestures, but non-manual features such as mouth movements, specifically mouthing, provide valuable linguistic information. This work directly classifies mouthing instances to their corresponding words in the spoken language while exploring the potential of transfer learning from Visual Speech Recognition (VSR) to mouthing recognition in German Sign Language. We leverage three VSR datasets: one in English, one in German with unrelated words and one in German containing the same target words as the mouthing dataset, to investigate the impact of task similarity in this setting. Our results demonstrate that multi-task learning improves both mouthing recognition and VSR accuracy as well as model robustness, suggesting that mouthing recognition should be treated as a distinct but related task to VSR. This research contributes to the field of SLR by proposing knowledge transfer from VSR to SLR datasets with limited mouthing annotations.

Letizia Vezzosi

In Present-Day English mutual situations are encoded either with lexical reciprocals or with reciprocal markers (each other). This was not the state of affairs in Middle English, which encodes mutual situation by means of both syntactic and morphological or clitic markers. The present paper will describe the inventory of morphological and syntactic devices of the Middle English reciprocal system in terms of semantic or syntactic constraints of usage, in order to shed light on its change into the typologically different Present-Day English reciprocal system. This development can be indeed explained in terms of a general restructuring of co-reference (reflexive and reciprocal) marking which took place in the course of the Middle English period.

Barbara Grobelna

This study has emerged from the research on telecinematic discourse and corpus linguistic studies. The aim of this paper is to provide a socio-pragmatic analysis of swear words in the dialogues of the Season 1 of the TV show “BoJack Horseman”, on the basis of the Bojack Horseman Corpus, prepared and compiled by the author. Lemmas such as crap, fuck, shit, damn, ass, dick, bitch, and suck are analysed, as well as their grammatical forms and functions they serve. The findings illustrate the importance of foul language employed by the show-writers and uncover the sociopragmatic aspect of the distribution of swear words in this particular TV show. The language used by the characters of the show depicts, to some extent real-life conversations, and swear words in them are used alike. The swear words in the dialogues function as an outlet for extreme emotions, a form of catharsis, or an extra emphasis added to the utterance. The offensiveness of swear words in “BoJack Horseman” is described as mild and moderate, but the swear words do occur in nearly every episode of the show. However, they are used not only to carry a negative impact (such as an insult or provocation), but also to convey positive information (the sense of social bonding or camaraderie).

Reet Bender

This article focuses on imaginative childhood games and fictional playmates in Baltic German children’s games, as seen in the context of, and in relation to, Baltic German domestic and educational culture. Through reminiscing about these games, a view of the Baltic German domestic environment in all its diversity is opened up. It gives an insight into the mental environment – relationships within the immediate and extended family and domestic circle (which also included servants) and the relationships and interactions beyond, but also shares information about the physical environment – the everyday culture and way of life. And in the midst of it all, of course, the narrators are looking back on their own formative experiences and years. The examples of more or less imaginative games are drawn from the memoirs of nine Baltic German authors, of which one text is available in Estonian translation, while the others are unknown to the Estonian reader as complete texts, and therefore perhaps all the more interesting. The authors have been selected from the corpus of Baltic German memoir texts collected from reading, based on the focus of this article, i.e., texts that also deal with the author’s childhood and where fictional playmates and worlds are also discussed. The selection is certainly not exhaustive, but it provides a preliminary insight into the subject. Baltic German memoir literature, especially concerning the generations whose childhoods passed during a period of stability and peace – whether at the end of the 19th century before Russification, or the pre-World War I era – often exhibits a particular contrast. Childhood and youth were left behind at a time when the world was still in order, when the life as it used to be had not collapsed. Later Baltic Germans very often wrote memoirs with their descendants in mind, in order to create a generational link, to introduce the lost world and way of life, but also to generalise the impact of historical events from the individual level to the national group as a whole. Post-World War II memory work directed at descendants became particularly significant, as everything had once again and irreversibly gone. The examples of imaginative childhood games and fictional playmates in the Baltic German memoirs discussed in this article carry a positive tone, occasionally accompanied by a light-hearted and wise reflection. Children were encouraged to engage in independent play, while not being neglected, especially the younger children, for whom the presence of an ‘interested adult’ in the background was noticeable. Storytelling and literature played a key role in the development of imagination and creativity. The plots of literary works were transferred to games and further developed, and dramatization techniques were employed using readily available resources. In addition to classical and romantic literature, the physical environment – buildings, interiors, the surrounding nature – also exerted important influence. Fictional companions existed both in the case of an only child and in the case of children growing up among siblings. In the latter case, there were often shared fantasy characters and worlds – whether they were just air or on paper. Adults also became part of the imaginary worlds and retained their role as interested observers – as long as the imaginary world did not overflow too much – leaving room for individual originality, which has been considered a distinctive feature of Baltic German society. The authors recall their childhood with fond memories, where home and family formed a stable, secure and vital growing environment, and the necessary resilience and resourcefulness for coping with life’s challenges. As such, these examples also fit into the framework of what Helene Hoerschelmann calls “old-school Baltic German upbringing”. Undoubtedly, the self-aware adaptability acquired during their childhood and youth served them well – the successive pivotal historical events, wars and revolutions became an inevitable part of their lives. Thus, a study of the history of the Baltic Germans’ childhood and upbringing could be of interest not only as part of an unknown or forgotten and lost world, but also as a multifaceted source of reflection and comparison, especially from today’s perspective.

Antonella Catone, Daniela Sorrentino

The present work aims at showing the didactic potential of the novel Zweinhalb Störche: Roman einer Kindheit in Siebenbürgen, written in 2008 by the German-Romanian author Claudiu M. Florian, for multilingual learning in GFL classes. The paper focuses on the possible use of corpus-based learning tasks and consists of two parts: the first will introduce the main features of multilingual didactics and multilingual competence and the possible teaching approaches of inter- and transcultural literature through the use of corpora; the second will focus on the use of the novel to promote multilingual competence within the GFL classroom. In particular, specific tasks will be suggested to achieve a didactic surplus through the use of corpus compilation and analysis tools such as Sketch Engine.

Wojciech Kunicki

Fynn Petersen-Frey, Chris Biemann

Extracting who says what to whom is a crucial part in analyzing human communication in today's abundance of data such as online news articles. Yet, the lack of annotated data for this task in German news articles severely limits the quality and usability of possible systems. To remedy this, we present a new, freely available, creative-commons-licensed dataset for quotation attribution in German news articles based on WIKINEWS. The dataset provides curated, high-quality annotations across 1000 documents (250,000 tokens) in a fine-grained annotation schema enabling various downstream uses for the dataset. The annotations not only specify who said what but also how, in which context, to whom and define the type of quotation. We specify our annotation schema, describe the creation of the dataset and provide a quantitative analysis. Further, we describe suitable evaluation metrics, apply two existing systems for quotation attribution, discuss their results to evaluate the utility of our dataset and outline use cases of our dataset in downstream tasks.

Yanran Chen, Wei Zhao, Anne Breitbarth et al.

Many studies have shown that human languages tend to optimize for lower complexity and increased communication efficiency. Syntactic dependency distance, which measures the linear distance between dependent words, is often considered a key indicator of language processing difficulty and working memory load. The current paper looks at diachronic trends in syntactic language change in both English and German, using corpora of parliamentary debates from the last c. 160 years. We base our observations on five dependency parsers, including the widely used Stanford CoreNLP as well as 4 newer alternatives. Our analysis of syntactic language change goes beyond linear dependency distance and explores 15 metrics relevant to dependency distance minimization (DDM) and/or based on tree graph properties, such as the tree height and degree variance. Even though we have evidence that recent parsers trained on modern treebanks are not heavily affected by data 'noise' such as spelling changes and OCR errors in our historic data, we find that results of syntactic language change are sensitive to the parsers involved, which is a caution against using a single parser for evaluating syntactic language change as done in previous work. We also show that syntactic language change over the time period investigated is largely similar between English and German for the different metrics explored: only 4% of cases we examine yield opposite conclusions regarding upwards and downtrends of syntactic metrics across German and English. We also show that changes in syntactic measures seem to be more frequent at the tails of sentence length distributions. To our best knowledge, ours is the most comprehensive analysis of syntactic language change using modern NLP technology in recent corpora of English and German.

Julian Linke, Saskia Wepner, Gernot Kubin et al.

As dialogue systems are becoming more and more interactional and social, also the accurate automatic speech recognition (ASR) of conversational speech is of increasing importance. This shifts the focus from short, spontaneous, task-oriented dialogues to the much higher complexity of casual face-to-face conversations. However, the collection and annotation of such conversations is a time-consuming process and data is sparse for this specific speaking style. This paper presents ASR experiments with read and conversational Austrian German as target. In order to deal with having only limited resources available for conversational German and, at the same time, with a large variation among speakers with respect to pronunciation characteristics, we improve a Kaldi-based ASR system by incorporating a (large) knowledge-based pronunciation lexicon, while exploring different data-based methods to restrict the number of pronunciation variants for each lexical entry. We achieve best WER of 0.4% on Austrian German read speech and best average WER of 48.5% on conversational speech. We find that by using our best pronunciation lexicon a similarly high performance can be achieved than by increasing the size of the data used for the language model by approx. 360% to 760%. Our findings indicate that for low-resource scenarios -- despite the general trend in speech technology towards using data-based methods only -- knowledge-based approaches are a successful, efficient method.

Regina Stodden, Omar Momen, Laura Kallmeyer

Text simplification is an intralingual translation task in which documents, or sentences of a complex source text are simplified for a target audience. The success of automatic text simplification systems is highly dependent on the quality of parallel data used for training and evaluation. To advance sentence simplification and document simplification in German, this paper presents DEplain, a new dataset of parallel, professionally written and manually aligned simplifications in plain German ("plain DE" or in German: "Einfache Sprache"). DEplain consists of a news domain (approx. 500 document pairs, approx. 13k sentence pairs) and a web-domain corpus (approx. 150 aligned documents, approx. 2k aligned sentence pairs). In addition, we are building a web harvester and experimenting with automatic alignment methods to facilitate the integration of non-aligned and to be published parallel documents. Using this approach, we are dynamically increasing the web domain corpus, so it is currently extended to approx. 750 document pairs and approx. 3.5k aligned sentence pairs. We show that using DEplain to train a transformer-based seq2seq text simplification model can achieve promising results. We make available the corpus, the adapted alignment methods for German, the web harvester and the trained models here: https://github.com/rstodden/DEPlain.

Michał Narajewski

The exponential growth of renewable energy capacity has brought much uncertainty to electricity prices and to electricity generation. To address this challenge, the energy exchanges have been developing further trading possibilities, especially the intraday and balancing markets. For an energy trader participating in both markets, the forecasting of imbalance prices is of particular interest. Therefore, in this manuscript we conduct a very short-term probabilistic forecasting of imbalance prices, contributing to the scarce literature in this novel subject. The forecasting is performed 30 minutes before the delivery, so that the trader might still choose the trading place. The distribution of the imbalance prices is modelled and forecasted using methods well-known in the electricity price forecasting literature: lasso with bootstrap, gamlss, and probabilistic neural networks. The methods are compared with a naive benchmark in a meaningful rolling window study. The results provide evidence of the efficiency between the intraday and balancing markets as the sophisticated methods do not substantially overperform the intraday continuous price index. On the other hand, they significantly improve the empirical coverage. The analysis was conducted on the German market, however it could be easily applied to any other market of similar structure.

Johann Petrak, Brigitte Krenn

This paper presents work on detecting misogyny in the comments of a large Austrian German language newspaper forum. We describe the creation of a corpus of 6600 comments which were annotated with 5 levels of misogyny. The forum moderators were involved as experts in the creation of the annotation guidelines and the annotation of the comments. We also describe the results of training transformer-based classification models for both binarized and original label classification of that corpus.

E. Teichmann, H. J. Lewandowski, M. Alemani

There is a large variety of goals instructors have for laboratory courses, with different courses focusing on different subsets of goals. An often implicit, but crucial, goal is to develop students` attitudes, views, and expectations about experimental physics to align with practicing experimental physicists. The assessment of laboratory courses upon this one dimension of learning has been intensively studied in US institutions using the Colorado Learning Attitudes Science Survey for Experimental Physics (E-CLASS). However, there is no such an instrument available to use in Germany, and the influence of laboratory courses on students` views about the nature of experimental physics is still unexplored at German-speaking institutions. Motivated by the lack of an assessment tool to investigate this goal in laboratory courses at German-speaking institutions, we present a translated version of the E-CLASS adapted to the context at German-speaking institutions. We call the German version of the E-CLASS, the GE-CLASS. We describe the translation process and the creation of an automated web-based system for instructors to assess their laboratory courses. We also present first results using GE-CLASS obtained at the University of Potsdam. A first comparison between E-CLASS and GE-CLASS results shows clear differences between University of Potsdam and US students' views and beliefs about experimental physics.