Datasets:

dgambettavuw

/

P_sciabs_doc1000_real96

like 0

Hate speech detection in online platforms has been widely studied inthe past. Most of these works were conducted in English and afew rich-resource languages. Recent approaches tailored forlow-resource languages have explored the interests of zero-shot cross-lingual transfer learning models in resource-scarce scenarios.However, languages variations between geolects such as AmericanEnglish and British English, Latin-American Spanish, and EuropeanSpanish is still a problem for NLP models that often relies on(latent) lexical information for their classification tasks. Moreimportantly, the cultural aspect, crucial for hate speech detection,is often overlooked.In this work, we present the results of a thorough analysis of hatespeech

Pre-trained models usually come with a pre-defined tokenization and little flexibility as to what subword tokens can be used in downstream tasks. This problem concerns especially multilingual NLP and low-resource languages, which are typically processed using cross-lingual transfer. In this paper, we aim to find out if the right granularity of tokenization is helpful for a text classification task, namely dialect classification. Aiming at generalizations beyond the studied cases, we look for the optimal granularity in four dialect datasets, two with relatively consistent writing (one Arabic and one Indo-Aryan set) and two with considerably inconsistent writing

One of the challenges with finetuning pretrained language models (PLMs) is that their tokenizer is optimized for the language(s) it was pretrained on, but brittle when it comes to previously unseen variations in the data. This can for instance be observed when finetuning PLMs on one language and evaluating them on data in a closely related language variety with no standardized orthography. Despite the high linguistic similarity, tokenization no longer corresponds to meaningful representations of the target data, leading to low performance in, e.g., part-of-speech tagging. In this work, we finetune PLMs on seven languages from

The digitisation of historical texts has provided new horizons for NLP research, but such data also presents a set of challenges, including scarcity and inconsistency. The lack of editorial standard during digitisation exacerbates these difficulties.This study explores the potential for temporal domain adaptation in Early Modern Irish and pre-reform Modern Irish data. We describe two experiments carried out on the book subcorpus of the Historical Irish Corpus, which includes Early Modern Irish and pre-reform Modern Irish texts from 1581 to 1926. We also propose a simple orthographic normalisation method for historical Irish that reduces the type-token

This paper measures variation in embedding spaces which have been trained on different regional varieties of English while controlling for instability in the embeddings. While previous work has shown that it is possible to distinguish between similar varieties of a language, this paper experiments with two follow-up questions: First, does the variety represented in the training data systematically influence the resulting embedding space after training? This paper shows that differences in embeddings across varieties are significantly higher than baseline instability. Second, is such dialect-based variation spread equally throughout the lexicon? This paper shows that specific parts

Automatic discrimination between Bosnian, Croatian, Montenegrin and Serbian is a hard task due to the mutual intelligibility of these South-Slavic languages. In this paper, we introduce the BENCHi{\'c}-lang benchmark for discriminating between these four languages. The benchmark consists of two datasets from different domains - a Twitter and a news dataset - selected with the aim of fostering cross-dataset evaluation of different modelling approaches. We experiment with the baseline SVM models, based on character n-grams, which perform nicely in-dataset, but do not generalize well in cross-dataset experiments. Thus, we introduce another approach, exploiting only web-crawled data

Eye-tracking data in Chinese languages present unique challenges due to the non-alphabetic and unspaced nature of the Chinese writing systems. This paper introduces the first deeply-annotated joint Mandarin-Cantonese eye-tracking dataset, from which we achieve a unified eye-tracking prediction system for both language varieties. In addition to the commonly studied first fixation duration and the total fixation duration, this dataset also includes the second fixation duration, expressing fixation patterns that are more relevant to higher-level, structural processing. A basic comparison of the features and measurements in our dataset revealed variation between Mandarin and Cantonese on fixation patterns

The study of low-resourced East Slavic lects is becoming increasingly relevant as they face the prospect of extinction under the pressure of standard Russian while being treated by academia as an inferior part of this lect. The Khislavichi lect, spoken in a settlement on the border of Russia and Belarus, is a perfect example of such an attitude.We take an alternative approach and study East Slavic lects (such as Khislavichi) as separate systems. The proposed method includes the development of a tagged corpus through morphological tagging with the models trained on the bigger lects. Morphological tagging

We introduce DiatopIt, the first corpus specifically focused on diatopic language variation in Italy for language varieties other than Standard Italian. DiatopIt comprises over 15K geolocated social media posts from Twitter over a period of two years, including regional Italian usage and content fully written in local language varieties or exhibiting code-switching with Standard Italian. We detail how we tackled key challenges in creating such a resource, including the absence of orthography standards for most local language varieties and the lack of reliable language identification tools. We assess the representativeness of DiatopIt across time and space,

The present work describes the solutions pro- posed by the UnibucNLP team to address the closed format of the DSL-TL task featured in the tenth VarDial Evaluation Campaign. The DSL-TL organizers provided approximately 11 thousand sentences written in three different languages and manually tagged with one of 9 classes. Out of these, 3 tags are considered common label and the remaining 6 tags are variety-specific. The DSL-TL task features 2 subtasks: Track 1 - a three-way and Track 2 - a two-way classification per language. In Track 2 only the variety-specific labels are used for scoring,

This paper addresses the need for massive corpora for a low-resource language and presents the publicly available UberText 2.0 corpus for the Ukrainian language and discusses the methodology of its construction. While the collection and maintenance of such a corpus is more of a data extraction and data engineering task, the corpus itself provides a solid foundation for natural language processing tasks. It can enable the creation of contemporary language models and word embeddings, resulting in a better performance of numerous downstream tasks for the Ukrainian language. In addition, the paper and software developed can be

This study addresses the challenges of learning unsupervised word representations for the morphologically rich and low-resource Ukrainian language. Traditional models that perform decently on English do not generalize well for such languages due to a lack of sufficient data and the complexity of their grammatical structures. To overcome these challenges, we utilized a high-quality, large dataset of different genres for learning Ukrainian word vector representations. We found the best hyperparameters to train fastText language models on this dataset and performed intrinsic and extrinsic evaluations of the generated word embeddings using the established methods and metrics. The

This paper presents an ongoing project to create the Ukrainian Brown Corpus (BRUK), a disambiguated corpus of Modern Ukrainian. Inspired by and loosely based on the original Brown University corpus, BRUK contains one million words, spans 11 years (2010{--}2020), and represents edited written Ukrainian. Using stratified random sampling, we have selected fragments of texts from multiple sources to ensure maximum variety, fill nine predefined categories, and produce a balanced corpus. BRUK has been automatically POS-tagged with the help of our tools (a large morphological dictionary of Ukrainian and a tagger). A manually disambiguated and validated subset

We present a corpus professionally annotated for grammatical error correction (GEC) and fluency edits in the Ukrainian language. We have built two versions of the corpus {--} GEC+Fluency and GEC-only {--} to differentiate the corpus application. To the best of our knowledge, this is the first GEC corpus for the Ukrainian language. We collected texts with errors (33,735 sentences) from a diverse pool of contributors, including both native and non-native speakers. The data cover a wide variety of writing domains, from text chats and essays to formal writing. Professional proofreaders corrected and annotated the corpus for

We present our system that participated in the shared task on the grammatical error correction of Ukrainian. We have implemented two approaches that make use of large pre-trained language models and synthetic data, that have been used for error correction of English as well as low-resource languages. The first approach is based on fine-tuning a large multilingual language model (mT5) in two stages: first, on synthetic data, and then on gold data. The second approach trains a (smaller) seq2seq Transformer model pre-trained on synthetic data and fine-tuned on gold data. Our mT5-based model scored first in

This article presents a comparative analysis of four different syntactic typological approaches applied to 20 different languages. We compared three specific quantitative methods, using parallel CoNLL-U corpora, to the classification obtained via syntactic features provided by a typological database (lang2vec). First, we analyzed the Marsagram linear approach which consists of extracting the frequency word-order patterns regarding the position of components inside syntactic nodes. The second approach considers the relative position of heads and dependents, and the third is based simply on the relative position of verbs and objects. From the results, it was possible to observe

Constituency parsing is an important task of informing how words are combined to form sentences. While constituency parsing in English has seen significant progress in the last few years, tools for constituency parsing in Indonesian remain few and far between. In this work, we publish ICON (Indonesian CONstituency treebank), the hitherto largest publicly-available manually-annotated benchmark constituency treebank for the Indonesian language with a size of 10,000 sentences and approximately 124,000 constituents and 182,000 tokens, which can support the training of state-of-the-art transformer-based models. We establish strong baselines on the ICON dataset using the Berkeley Neural Parser

Historical treebanking within the generative framework has gained in popularity. However, there are still many languages and historical periods yet to be represented. For German, a constituency treebank exists for historical Low German, but not Early New High German. We begin to fill this gap by presenting our initial work on the Parsed Corpus of Early New High German (PCENHG). We present the methodological considerations and workflow for the treebank{'}s annotations and development. Given the limited amount of currently available PCENHG treebank data, we treat it as a low-resource language and leverage a larger, closely related

Sound correspondence patterns form the basis of cognate detection and phonological reconstruction in historical language comparison. Methods for the automatic inference of correspondence patterns from phonetically aligned cognate sets have been proposed, but their application to multilingual wordlists requires extremely well annotated datasets. Since annotation is tedious and time consuming, it would be desirable to find ways to improve aligned cognate data automatically. Taking inspiration from trimming techniques in evolutionary biology, which improve alignments by excluding problematic sites, we propose a workflow that trims phonetic alignments in comparative linguistics prior to the inference of correspondence patterns.

The success of cross-lingual transfer learning for POS tagging has been shown to be strongly dependent, among other factors, on the (typological and/or genetic) similarity of the low-resource language used for testing and the language(s) used in pre-training or to fine-tune the model. We further unpack this finding in two directions by zooming in on a single language, namely Persian. First, still focusing on POS tagging we run an in-depth analysis of the behaviour of Persian with respect to closely related languages and languages that appear to benefit from cross-lingual transfer with Persian. To do so,

We present a cross-linguistic study of vowel harmony that aims to quantifies this phenomenon using data-driven computational modeling. Concretely, we define an information-theoretic measure of harmonicity based on the predictability of vowels in a natural language lexicon, which we estimate using phoneme-level language models (PLMs). Prior quantitative studies have heavily relied on inflected word-forms in the analysis on vowel harmony. On the contrary, we train our models using cross-linguistically comparable lemma forms with little or no inflection, which enables us to cover more under-studied languages. Training data for our PLMs consists of word lists offering a

In this replication study of previous research into dependency length minimisation (DLM), we pilot a new parallel multilingual parsed corpus to examine whether previous findings are upheld when controlling for variation in domain and sentence content between languages. We follow the approach of previous research in comparing the dependency lengths of observed sentences in a multilingual corpus to a variety of baselines: permutations of the sentences, either random or according to some fixed schema. We go on to compare DLM with intervener complexity measure (ICM), an alternative measure of syntactic complexity. Our findings uphold both dependency

While multilingual language models can improve NLP performance on low-resource languages by leveraging higher-resource languages, they also reduce average performance on all languages (the {`}curse of multilinguality{'}). Here we show another problem with multilingual models: grammatical structures in higher-resource languages bleed into lower-resource languages, a phenomenon we call grammatical structure bias. We show this bias via a novel method for comparing the fluency of multilingual models to the fluency of monolingual Spanish and Greek models: testing their preference for two carefully-chosen variable grammatical structures (optional pronoun-drop in Spanish and optional Subject-Verb ordering in Greek). We find

In morphology, a distinction is commonly drawn between inflection and derivation. However, a precise definition of this distinction which captures the way the terms are used across languages remains elusive within linguistic theory, typically being based on subjective tests. In this study, we present 4 quantitative measures which use the statistics of a raw text corpus in a language to estimate how much and how variably a morphological construction changes aspects of the lexical entry, specifically, the word{'}s form and the word{'}s semantic and syntactic properties (as operationalised by distributional word embeddings). Based on a sample

Transformer-based language models (LMs) offer superior performance in a wide range of NLP tasks compared to previous paradigms. However, the vast majority of the world{'}s languages do not have adequate training data available for monolingual LMs (Joshi et al., 2020). While the use of multilingual LMs might address this data imbalance, there is evidence that multilingual LMs struggle when it comes to model adaptation to to resource-poor languages (Wu and Dredze, 2020), or to languages which have typological characteristics unseen by the LM ({\"U}st{\"u}n et al., 2022). Other approaches aim to adapt monolingual LMs to resource-poor

Self-supervision has emerged as an effective paradigm for learning representations of spoken language from raw audio without explicit labels or transcriptions. Self-supervised speech models, such as wav2vec 2.0 (Baevski et al., 2020) and HuBERT (Hsu et al., 2021), have shown significant promise in improving the performance across different speech processing tasks. One of the main advantages of self-supervised speech models is that they can be pre-trained on a large sample of languages (Conneau et al., 2020; Babu et al.,2022), which facilitates cross-lingual transfer for low-resource languages (San et al., 2021). State-of-the-art self-supervised speech models include a

In this paper, we present the newly compiled DA-ELEXIS Corpus, which is one of the largest sense-annotated corpora available for Danish, and the first one to be annotated with the Danish wordnet, DanNet. The corpus is part of a European initiative, the ELEXIS project, and has corresponding parallel annotations in nine other European languages. As such it functions as a cross-lingual evaluative benchmark for a series of low and medium resourced European language. We focus here on the Danish annotation process, i.e. on the annotation scheme including annotation guidelines and a primary sense inventory constituted by

In this paper, we present an approach for translating word embeddings from a majority language into 4 minority languages: Erzya, Moksha, Udmurt and Komi-Zyrian. Furthermore, we align these word embeddings and present a novel neural network model that is trained on English data to conduct sentiment analysis and then applied on endangered language data through the aligned word embeddings. To test our model, we annotated a small sentiment analysis corpus for the 4 endangered languages and Finnish. Our method reached at least 56{\%} accuracy for each endangered language. The models and the sentiment corpus will be

The developments of deep learning in natural language processing (NLP) in recent years have resulted in an unprecedented amount of computational power and data required to train state-of-the-art NLP models. This makes lower-resource languages, such as Swedish, increasingly more reliant on language transfer effects from English since they do not have enough data to train separate monolingual models. In this study, we investigate whether there is any potential loss in English-Swedish language transfer by evaluating two types of language transfer on the GLUE/SweDiagnostics datasets and comparing between different linguistic phenomena. The results show that for an

Negation constitutes a challenging phenomenon for many natural language processing tasks, such as sentiment analysis (SA). In this paper we investigate the relationship between negation and sentiment in the context of Norwegian professional reviews. The first part of this paper includes a corpus study which investigates how negation is tied to sentiment in this domain, based on existing annotations. In the second part, we introduce NoReC-NegSynt, a synthetically augmented test set for negation and sentiment, to allow for a more detailed analysis of the role of negation in current neural SA models. This diagnostic test set,

Large language models (LLMs) have substantially improved natural language processing (NLP) performance, but training these models from scratch is resource-intensive and challenging for smaller languages.With this paper, we want to initiate a discussion on the necessity of language-specific pre-training of LLMs.We propose how the {``}one model-many models{''} conceptual framework for task transfer can be applied to language transfer and explore this approach by evaluating the performance of non-Swedish monolingual and multilingual models{'} performance on tasks in Swedish.Our findings demonstrate that LLMs exposed to limited Swedish during training can be highly capable and transfer competencies from English

This paper presents a spell checker and correction tool specifically designed for Wolof, an under-represented spoken language in Africa. The proposed spell checker leverages a combination of a trie data structure, dynamic programming, and the weighted Levenshtein distance to generate suggestions for misspelled words. We created novel linguistic resources for Wolof, such as a lexicon and a corpus of misspelled words, using a semi-automatic approach that combines manual and automatic annotation methods. Despite the limited data available for the Wolof language, the spell checker{'}s performance showed a predictive accuracy of 98.31{\%} and a suggestion accuracy of

In this study, we investigate the effectiveness of using cross-lingual word embeddings for zero-shot transfer learning between a language with an abundant resource, English, and a languagewith limited resource, isiZulu. IsiZulu is a part of the South African Nguni language family, which is characterised by complex agglutinating morphology. We use VecMap, an open source tool, to obtain cross-lingual word embeddings. To perform an extrinsic evaluation of the effectiveness of the embeddings, we train a news classifier on labelled English data in order to categorise unlabelled isiZulu data using zero-shot transfer learning. In our study, we found

This paper introduces two multilingual government themed corpora in various South African languages. The corpora were collected by gathering South African government speeches (ZA-gov-multilingual), as well as the South African Government newspaper (Vuk{'}uzenzele), that are translated into all 11 South African official languages. The corpora can be used for a myriad of downstream NLP tasks. The corpora were created to allow researchers to study the language used in South African government publications, with a focus on understanding how South African government officials communicate with their constituents. In this paper we highlight the process of gathering, cleaning

In this paper we present a case study for three under-resourced linguistically distinct South African languages (Afrikaans, isiZulu, and Sesotho sa Leboa) to investigate the influence of data size and linguistic nature of a language on the performance of different embedding types. Our experimental setup consists of training embeddings on increasing amounts of data and then evaluating the impact of data size for the downstream task of part of speech tagging. We find that relatively little data can produce useful representations for this specific task for all three languages. Our analysis also shows that the influence

The purpose of this article is to demonstrate that the recently developed automated rule-based syllabification system for Sesotho can be used broadly across the officially recognised South African Sotho-Tswana language group encompassing Sepedi, Sesotho and Setswana. We evaluate the automatic syllabification system on 400 words comprising 100 most frequently used words and 100 least-used words in Sepedi and Setswana as evident in the Autshumato corpus publicly available online. It is found that the Sesotho rule-based syllabification system can be used to correctly identify vowel-only syllables, consonant-vowel syllables and consonant-only syllables in Sepedi and Setswana. Among other

Orthographical standardization is a milestone in a language{'}s documentation and the development of its resources. However, texts written in former orthographies remain relevant to the language{'}s history and development and therefore must be converted to the standardized orthography. Ensuring a language has access to the orthographically standardized version of all of its recorded texts is important in the development of resources as it provides additional textual resources for training, supports contribution of authors using former writing systems, and provides information about the development of the language. This paper evaluates the performance of natural language processing methods,

We investigate to what extent pre-trained language models can be used for automated claim detection for fact-checking in a low resource setting. We explore this idea by fine-tuning four Norwegian pre-trained language models to perform the binary classification task of determining if a claim should be discarded or upheld to be further processed by human fact-checkers. We conduct a set of experiments to compare the performance of the language models, and provide a simple baseline model using SVM with tf-idf features. Since we are focusing on claim detection, the recall score for the \textit{upheld} class is

Data anonymisation is often required to comply with regulations when transfering information across departments or entities. However, the risk is that this procedure can distort the data and jeopardise the models built on it. Intuitively, the process of training an NLP model on anonymised data may lower the performance of the resulting model when compared to a model trained on non-anonymised data. In this paper, we investigate the impact of de-identification on the performance of nine downstream NLP tasks. We focus on the anonymisation and pseudonymisation of personal names and compare six different anonymisation strategies for

In this work, we studied methods for automatic abstractive summarization in a low-resource setting using Icelandic text, which is morphologically rich and has limited data compared to languages such as English. We collected and published the first publicly available abstractive summarization dataset for Icelandic and used it for training and evaluation of our models. We found that using multilingual pre-training in this setting led to improved performance, with the multilingual mT5 model consistently outperforming a similar model pre-trained from scratch on Icelandic text only. Additionally, we explored the use of machine translations for fine-tuning data augmentation

We train and evaluate four Part-of-Speech tagging models for Icelandic. Three are older models that obtained the highest accuracy for Icelandic when they were introduced. The fourth model is of a type that currently reaches state-of-the-art accuracy. We use the most recent version of the MIM-GOLD training/testing corpus, its newest tagset, and augmentation data to obtain results that are comparable between the various models. We examine the accuracy improvements with each model and analyse the errors produced by our transformer model, which is based on a previously published ConvBERT model. For the set of errors that

This work studies the plausibility of sequence-to-sequence neural networks as models of morphological acquisition by humans. We replicate the findings of Kirov and Cotterell (2018) on the well-known challenge of the English past tense and examine their generalizability to two related but morphologically richer languages, namely Dutch and German. Using a new dataset of English/Dutch/German (ir)regular verb forms, we show that the major findings of Kirov and Cotterell (2018) hold for all three languages, including the observation of over-regularization errors and micro U-shape learning trajectories. At the same time, we observe troublesome cases of non human-like

We study the performance of machine learning techniques to the problem of identifying speakers at meetings from anonymous minutes issued afterwards. The data comes from board meetings of Sveriges Riksbank (Sweden{'}s Central Bank). The data is split in two ways, one where each reported contribution to the discussion is treated as a data point, and another where all contributions from a single speaker have been aggregated. Using interpretable models we find that lexical features and topic models generated from speeches held by the board members outside of board meetings are good predictors of speaker identity. Combining

Dialectal variation is present in many human languages and is attracting a growing interest in NLP. Most previous work concentrated on either (1) classifying dialectal varieties at the document or sentence level or (2) performing standard NLP tasks on dialectal data. In this paper, we propose the novel task of token-level dialectal feature prediction. We present a set of fine-grained annotation guidelines for Norwegian dialects, expand a corpus of dialectal tweets, and manually annotate them using the introduced guidelines. Furthermore, to evaluate the learnability of our task, we conduct labeling experiments using a collection of baselines,

This paper introduces a Scandinavian benchmarking platform, ScandEval, which can benchmark any pretrained model on four different tasks in the Scandinavian languages. The datasets used in two of the tasks, linguistic acceptability and question answering, are new. We develop and release a Python package and command-line interface, scandeval, which can benchmark any model that has been uploaded to the Hugging Face Hub, with reproducible results. Using this package, we benchmark more than 80 Scandinavian or multilingual models and present the results of these in an interactive online leaderboard, as well as provide an analysis of the

Retrieval-based language models are increasingly employed in question-answering tasks. These models search in a corpus of documents for relevant information instead of having all factual knowledge stored in its parameters, thereby enhancing efficiency, transparency, and adaptability. We develop the first Norwegian retrieval-based model by adapting the REALM framework and evaluate it on various tasks. After training, we also separate the language model, which we call the \textit{reader}, from the retriever components, and show that this can be fine-tuned on a range of downstream tasks. Results show that retrieval augmented language modeling improves the reader{'}s performance on

Machine Translation (MT) has become an integral part of daily life for millions of people, with its output being so fluent that users often cannot distinguish it from human translation. However, these fluid texts often harbor algorithmic traces, from limited lexical choices to societal misrepresentations. This raises concerns about the possible effects of MT on natural language and human communication and calls for regular evaluations of machine-generated translations for different languages. Our paper explores the output of three widely used engines (Google, DeepL, Microsoft Azure) and one smaller commercial system. We translate the English and French

There are several ways of implementing multilingual NLP systems but little consensus as to whether different approaches exhibit similar effects. Are the trends that we observe when adding more languages the same as those we observe when sharing more parameters? We focus on encoder representations drawn from modular multilingual machine translation systems in an English-centric scenario, and study their quality from multiple aspects: how adequate they are for machine translation, how independent of the source language they are, and what semantic information they convey. Adding translation directions in English-centric scenarios does not conclusively lead to an

Automatic abstractive text summarization is a challenging task in the field of natural language processing. This paper presents a model for domain-specific sum marization for Danish news articles, Dan SumT5; an mT5 model fine-tuned on a cleaned subset of the DaNewsroom dataset consisting of abstractive summary-article pairs. The resulting state-of-the-art model is evaluated both quantitatively and qualitatively, using ROUGE and BERTScore metrics and human rankings of the summaries. We find that although model refinements increase quantitative and qualitative performance, the model is still prone to factual errors. We discuss the limitations of current evaluation methods for

Learning a new language is often difficult, especially practising it independently. The main issue with self-study is the absence of accurate feedback from a teacher, which would enable students to learn unfamiliar languages. In recent years, with advances in Artificial Intelligence and Automatic Speech Recognition, it has become possible to build applications that can provide valuable feedback on the users{'} pronunciation. In this paper, we introduce the CaptainA app explicitly developed to aid students in practising their Finnish pronunciation on handheld devices. Our app is a valuable resource for immigrants who are busy with school or

Sentiment classification is valuable for literary analysis, as sentiment is crucial in literary narratives. It can, for example, be used to investigate a hypothesis in the literary analysis of 19th-century Scandinavian novels that the writing of female authors in this period was characterized by negative sentiment, as this paper shows. In order to enable a data-driven analysis of this hypothesis, we create a manually annotated dataset of sentence-level sentiment annotations for novels from this period and use it to train and evaluate various sentiment classification methods. We find that pre-trained multilingual language models outperform models trained

Despite much progress in recent years, the vast majority of work in natural language processing (NLP) is on standard languages with many speakers. In this work, we instead focus on low-resource languages and in particular non-standardized low-resource languages. Even within branches of major language families, often considered well-researched, little is known about the extent and type of available resources and what the major NLP challenges are for these language varieties. The first step to address this situation is a systematic survey of available corpora (most importantly, annotated corpora, which are particularly valuable for NLP research). Focusing

We present a method for analyzing character errors for use with character-based, end-to-end ASR systems, as used herein for investigating dialectal speech. As end-to-end systems are able to produce novel spellings, there exists a possibility that the spelling variants produced by these systems can capture phonological information beyond the intended target word. We therefore first introduce a way of guaranteeing that similar words and characters are paired during alignment, thus ensuring that any resulting analysis of character errors is founded on sound substitutions. Then, from such a careful character alignment, we find trends in system-generated spellings

This paper describes the creation and evaluation of a synthetic dataset of Swedish multiple-choice questions (MCQs) for reading comprehension using GPT-3. Although GPT-3 is trained mostly on English data, with only 0.11{\%} of Swedish texts in its training material, the model still managed to generate MCQs in Swedish. About 44{\%} of the generated MCQs turned out to be of sufficient quality, i.e. they were grammatically correct and relevant, with exactly one answer alternative being correct and the others being plausible but wrong. We provide a detailed analysis of the errors and shortcomings of the rejected MCQs,

With large amounts of training data, it is possible to train ASR models that generalize well across speakers and domains. But how do you train robust models when there is a limited amount of available training data? In the experiments reported here, we fine-tuned a pre-trained wav2vec2 ASR model on two transcribed, Norwegian speech datasets, one with parliamentary speech and one with radio recordings, as well as on combinations of the two datasets. We subsequently tested these models on different test sets with planned and unplanned speech and with speakers of various dialects. Our results show

Recent advances in the field of language modeling have improved the state-of-the-art in question answering (QA) and question generation (QG). However, the development of modern neural models, their benchmarks, and datasets for training them has mainly focused on English. Finnish, like many other languages, faces a shortage of large QA/QG model training resources, which has prevented experimenting with state-of-the-art QA/QG fine-tuning methods. We present the first neural QA and QG models that work with Finnish. To train the models, we automatically translate the SQuAD dataset and then use normalization methods to reduce the amount of problematic

Contradictory results about the encoding of the semantic impact of negation in pretrained language models (PLMs) have been drawn recently (e.g. Kassner and Sch{\"u}tze (2020); Gubelmann and Handschuh (2022)).In this paper we focus rather on the way PLMs encode negation and its formal impact, through the phenomenon of the Negative Polarity Item (NPI) licensing in English.More precisely, we use probes to identify which contextual representations best encode 1) the presence of negation in a sentence, and 2) the polarity of a neighboring masked polarity item.We find that contextual representations of tokens inside the negation scope do

We present Gamli, an ASR corpus for Icelandic oral histories, the first of its kind for this language, derived from the {\'I}sm{\'u}s ethnographic collection. Corpora for oral histories differ in various ways from corpora for general ASR, they contain spontaneous speech, multiple speakers per channel, noisy environments, the effects of historic recording equipment, and typically a large proportion of elderly speakers. Gamli contains 146 hours of aligned speech and transcripts, split into a training set and a test set. We describe our approach for creating the transcripts, through both OCR of previous transcripts and post-editing of

In recent years, instruction finetuning models have received increased attention due to their remarkable zero-shot and generalization capabilities. However, the widespread implementation of these models has been limited to the English language, largely due to the costs and challenges associated with creating instruction datasets. To overcome this, automatic instruction generation has been proposed as a resourceful alternative. We see this as an opportunity for the adoption of instruction finetuning for other languages.In this paper we explore the viability of instruction finetuning for Swedish. We translate a dataset of generated instructions from English to Swedish, using it

Long term language technology infrastructures are critical for continued maintenance of language technology based software that is used to support the use of languages in digital world. In Nordic area we have languages ranging from well-resourced national majority languages like Norwegian, Swedish and Finnish as well as minoritised, unresourced and indigenous languages like S{\'a}mi languages. We present an infrastructure that has been build in over 20 years time that supports building language technology and tools for most of the Nordic languages as well as many of the languages all over the world, with focus on S{\'a}mi

In semantic typology, colexification refers to words with multiple meanings, either related (polysemy) or unrelated (homophony). Studies of cross-linguistic colexification have yielded insights into, e.g., psychology, historical linguistics and cognitive science (Xu et al., 2020; Brochhagen and Boleda, 2022; Schapper and Koptjevskaja-Tamm, 2022). While NLP research up until now has mainly focused on integrating syntactic typology (Naseem et al., 2012; Ponti et al., 2019; Chaudhary et al., 2019; {\"U}st{\"u}n et al., 2020; Ansell et al., 2021; Oncevay et al., 2022), we here investigate the potential of incorporating semantic typology, of which colexification is an example. We

Due to the popularity of social media platforms and the sheer amount of user-generated content online, the automatic detection of toxic language has become crucial in the creation of a friendly and safe digital space. Previous work has been mostly focusing on English leaving many lower-resource languages behind. In this paper, we present novel resources for toxicity detection in Finnish by introducing two new datasets, a machine translated toxicity dataset for Finnish based on the widely used English Jigsaw dataset and a smaller test set of Suomi24 discussion forum comments originally written in Finnish and manually

Multilingual language models have pushed state-of-the-art in cross-lingual NLP transfer. The majority of zero-shot cross-lingual transfer, however, use one and the same massively multilingual transformer (e.g., mBERT or XLM-R) to transfer to all target languages, irrespective of their typological, etymological, and phylogenetic relations to other languages. In particular, readily available data and models of resource-rich sibling languages are often ignored. In this work, we empirically show, in a case study for Faroese {--} a low-resource language from a high-resource language family {--} that by leveraging the phylogenetic information and departing from the {`}one-size-fits-all{'} paradigm, one can

This paper presents the annotation process of two Estonian named entity recognition (NER) datasets, involving the creation of annotation guidelines for labeling eleven different types of entities. In addition to the commonly annotated entities such as person names, organization names, and locations, the annotation scheme encompasses geopolitical entities, product names, titles/roles, events, dates, times, monetary values, and percents. The annotation was performed on two datasets, one involving reannotating an existing NER dataset primarily composed of news texts and the other incorporating new texts from news and social media domains. Transformer-based models were trained on these annotated

The identification of Verbal Multiword Expressions (VMWEs) presents a greater challenge compared to non-verbal MWEs due to their higher surface variability. VMWEs are linguistic units that exhibit varying levels of semantic opaqueness and pose difficulties for computational models in terms of both their identification and the degree of compositionality. In this study, a new approach to predicting the compositional nature of VMWEs in Persian is presented. The method begins with an automatic identification of VMWEs in Persian sentences, which is approached as a sequence labeling problem for recognizing the components of VMWEs. The method then creates

This paper shines a light on the potential of definition-based semantic models for detecting idiomatic and semi-idiomatic multiword expressions (MWEs) in clinical terminology. Our study focuses on biomedical entities defined in the UMLS ontology and aims to help prioritize the translation efforts of these entities. In particular, we develop an effective tool for scoring the idiomaticity of biomedical MWEs based on the degree of similarity between the semantic representations of those MWEs and a weighted average of the representation of their constituents. We achieve this using a biomedical language model trained to produce similar representations for

In many humanitarian scenarios, translation into severely low resource languages often does not require a universal translation engine, but a dedicated text-specific translation engine. For example, healthcare records, hygienic procedures, government communication, emergency procedures and religious texts are all limited texts. While generic translation engines for all languages do not exist, translation of multilingually known limited texts into new, endangered languages may be possible and reduce human translation effort. We attempt to leverage translation resources from rich resource languages to efficiently produce best possible translation quality for well known texts, which is available in multiple languages,

We propose a two-stage approach for training a single NMT model to translate unseen languages both to and from English. For the first stage, we initialize an encoder-decoder model to pretrained XLM-R and RoBERTa weights, then perform multilingual fine-tuning on parallel data in 40 languages to English. We find this model can generalize to zero-shot translations on unseen languages. For the second stage, we leverage this generalization ability to generate synthetic parallel data from monolingual datasets, then bidirectionally train with successive rounds of back-translation.Our approach, which we EcXTra (uE/unglish-uc/uentric Crosslingual (uX/u) uTra/unsfer), is conceptually simple, only

Multilingual pre-training significantly improves many multilingual NLP tasks, including machine translation. Most existing methods are based on some variants of masked language modeling and text-denoising objectives on monolingual data. Multilingual pre-training on monolingual data ignores the availability of parallel data in many language pairs. Also, some other works integrate the available human-generated parallel translation data in their pre-training. This kind of parallel data is definitely helpful, but it is limited even in high-resource language pairs. This paper introduces a novel semi-supervised method, SPDG, that generates high-quality pseudo-parallel data for multilingual pre-training. First, a denoising model is

Large multilingual models trained with self-supervision achieve state-of-the-art results in a wide range of natural language processing tasks. Self-supervised pretrained models are often fine-tuned on parallel data from one or multiple language pairs for machine translation. Multilingual fine-tuning improves performance on low-resource languages but requires modifying the entire model and can be prohibitively expensive. Training a new adapter on each language pair or training a single adapter on all language pairs without updating the pretrained model has been proposed as a parameter-efficient alternative. However, the former does not permit any sharing between languages, while the latter

Machine translation (MT) involving Indigenous languages, including endangered ones, is challenging primarily due to lack of sufficient parallel data. We describe an approach exploiting bilingual and multilingual pretrained MT models in a transfer learning setting to translate from Spanish into ten South American Indigenous languages. Our models set new SOTA on five out of the ten language pairs we consider, even doubling performance on one of these five pairs. Unlike previous SOTA that perform data augmentation to enlarge the train sets, we retain the low-resource setting to test the effectiveness of our models under such a

Data sparsity is a main problem hindering the development of code-switching (CS) NLP systems. In this paper, we investigate data augmentation techniques for synthesizing dialectal Arabic-English CS text. We perform lexical replacements using word-aligned parallel corpora where CS points are either randomly chosen or learnt using a sequence-to-sequence model. We compare these approaches against dictionary-based replacements. We assess the quality of generated sentences through human evaluation and evaluate the effectiveness of data augmentation on machine translation (MT), automatic speech recognition (ASR), and speech translation (ST) tasks. Results show that using a predictive model results in more

Data augmentation (DA) is a popular strategy to boost performance on neural machine translation tasks. The impact of data augmentation in low-resource environments, particularly for diverse and scarce languages, is understudied. In this paper, we introduce a simple yet novel metric to measure the impact of several different data augmentation strategies. This metric, which we call Data Augmentation Advantage (DAA), quantifies how many true data pairs a synthetic data pair is worth in a particular experimental context. We demonstrate the utility of this metric by training models for several linguistically-varied datasets using the data augmentation methods

Parallel corpora are still crucial to train effective Machine Translation systems. This is even more true for low-resource language pairs, for which Neural Machine Translation has been shown to be less robust to domain mismatch and noise. Due to time and resource constraints, parallel corpora are mostly created with sentence alignment methods which automatically infer alignments. Recent work focused on state-of-the-art pre-trained sentence embeddings-based methods which are available only for a tiny fraction of the world{'}s languages. In this paper, we evaluate the performance of four widely used algorithms on the low-resource English-Yor{\`u}b{\'a} language pair against

This paper investigates the use of standard and non-standard adverbial markers in modern Chinese literature. In Chinese, adverbials can be derived from many adjectives, adverbs and verbs with the suffix {``}de{''}. The suffix has a standard and a non-standard written form, both of which are frequently used. Contrastive research on these two competing forms has mostly been qualitative or limited to small text samples. In this first large-scale quantitative study, we present statistics on 346 adverbial types from an 8-million-character text corpus drawn from Chinese literature in the 20th century. We present a semantic analysis of

Poem generation with language models requires the modeling of rhyming patterns. We propose a novel solution for learning to rhyme, based on synthetic data generated with a rule-based rhyming algorithm. The algorithm and an evaluation metric use a phonetic dictionary and the definitions of perfect and assonant rhymes. We fine-tune a GPT-2 English model with 124M parameters on 142 MB of natural poems and find that this model generates consecutive rhymes infrequently (11{\%}). We then fine-tune the model on 6 MB of synthetic quatrains with consecutive rhymes (AABB) and obtain nearly 60{\%} of rhyming lines in

Quotes are universally appealing. Humans recognize good quotes and save them for later reference. However, it may pose a challenge for machines. In this work, we build a new corpus of quotes and propose a new task, quote detection, as a type of span detection. We retrieve the quote set from Goodreads and collect the spans through a custom search on the Gutenberg Book Corpus. We measure unique vocabulary usage by a state-of-the-art language model and perform comparative statistical analysis against the Cornell Movie-Quotes Corpus. Furthermore, we run two types of baselines for quote detection: Conditional

We present a dataset and system for quote attribution in Dutch literature. The system is implemented as a neural module in an existing NLP pipeline for Dutch literature (dutchcoref; van Cranenburgh, 2019). Our contributions are as follows. First, we provide guidelines for Dutch quote attribution and annotate 3,056 quotes in fragments of 42 Dutch literary novels, both contemporary and classic. Second, we present three neural quote attribution classifiers, optimizing for precision, recall, and F1. Third, we perform an evaluation and analysis of quote attribution performance, showing that in particular, quotes with an implicit speaker are challenging,

The subject of this article is the application of NLP and text-mining methods to the analysis of two large bibliographies: Polish one, based on the catalogs of the National Library in Warsaw, and the other German one, created by Deutsche Nationalbibliothek. The data in both collections are stored in MARC 21 format, allowing the selection of relevant fields that are used for further processing (basically author, title, and date). The volume of the Polish corpus (after filtering out non-relevant or incomplete items) includes 1.4 mln of records, and that of the German corpus 7.5 mln records.

Emotion regulation is a crucial element in dealing with emotional events and has positive effects on mental health. This paper aims to provide a more comprehensive understanding of emotional events by introducing a new French corpus of emotional narratives collected using a questionnaire for emotion regulation. We follow the theoretical framework of the Component Process Model which considers emotions as dynamic processes composed of four interrelated components (behavior, feeling, thinking and territory). Each narrative is related to a discrete emotion and is structured based on all emotion components by the writers. We study the interaction of

Development funds are essential to finance climate change adaptation and are thus an important part of international climate policy. How ever, the absence of a common reporting practice makes it difficult to assess the amount and distribution of such funds. Research has questioned the credibility of reported figures, indicating that adaptation financing is in fact lower than published figures suggest. Projects claiming a greater relevance to climate change adaptation than they target are referred to as {``}overreported{''}. To estimate realistic rates of overreporting in large data sets over times, we propose an approach based on state-of-the-art

The presence of bias is a pressing concern for both engineers and users of language technology. What is less clear is how exactly bias can be measured, so as to rank models relative to the biases they display. Using an innovative experimental method involving data augmentation, we measure the effect of intersectional biases in Danish models used for Name Entity Recognition (NER). We quantify differences in representational biases, understood as a systematic difference in error or what is called error disparity. Our analysis includes both gender and ethnicity to illustrate the effect of multiple dimensions of

Olfaction is a rather understudied sense compared to the other senses. In NLP, however, there have been recent attempts to develop taxonomies and benchmarks specifically designed to capture smell-related information. In this work, we further extend this research line by presenting a supervised system for olfactory information extraction in English. We cast this problem as a token classification task and build a system that identifies smell words, smell sources and qualities. The classifier is then applied to a set of English historical corpora, covering different domains and written in a time period between the 15th and

This study relies on natural language processing to explore the nature of online communication in Russia during the war on Ukraine in 2022. The analysis of a large corpus of publications in traditional media and on social media identifies massive state interventions aimed at manipulating public opinion. The study relies on expertise in media studies and political science to trace the major themes and strategies of the propagandist narratives on three major Russian social media platforms over several months as well as their perception by the users. Distributions of several keyworded pro-war and anti-war topics are

In social sciences, recent years have witnessed a growing interest in applying NLP approaches to automatically detect framing in political discourse. However, most NLP studies by now focus heavily on framing effect arising from topic coverage, whereas framing effect arising from subtle usage of linguistic devices remains understudied. In a collaboration with political science researchers, we intend to investigate framing strategies in German newspaper articles on the {``}European Refugee Crisis{''}. With the goal of a more in-depth framing analysis, we not only incorporate lexical cues for shallow topic-related framing, but also propose and operationalize a variety

In this study, we demonstrate how to apply cross-lingual annotation projection to transfer named-entity annotations to classical languages for which limited or no resources and annotated texts are available, aiming to enrich their NER training datasets and train a model to perform NER tagging. Our method uses sentence-level aligned parallel corpora ancient texts and the translation in a modern language, for which high-quality off-the-shelf NER systems are available. We automatically annotate the text of the modern language and employ a state-of-the-art neural word alignment system to find translation equivalents. Finally, we transfer the annotations to the

Embedding representations of text are useful for downstream natural language processing tasks. Several universal sentence representation methods have been proposed with a particular focus on self-supervised pre-training approaches to leverage the vast quantities of unlabelled data. However, there are two challenges for generating rich embedding representations for a new document. 1) The latest rich embedding generators are based on very large costly transformer-based architectures. 2) The rich embedding representation of a new document is limited to only the information provided without access to any explicit contextual and temporal information that could potentially further enrich the representation.

In recent years, there has been growing interest in the field of abstractive text summarization with focused contributions in relevant model architectures, datasets, and evaluation metrics. Despite notable research advances, previous works have identified certain limitations concerning the quality of datasets and the effectiveness of evaluation techniques for generated summaries. In this context, we examine these limitations further with the help of three quality measures, namely, Information Coverage, Entity Hallucination, and Summarization Complexity. As a part of this work, we investigate two widely used datasets (XSUM and CNNDM) and three existing models (BART, PEGASUS, and BRIO)

In this paper, we describe how we unearthed some fundamental problems while building an analogy dataset modelled on BATS (Gladkova et al., 2016) to evaluate historical Irish embeddings on their ability to detect orthographic, morphological and semantic similarity.The performance of our models in the analogy task was extremely poor regardless of the architecture, hyperparameters and evaluation metrics, while the qualitative evaluation revealed positive tendencies. We argue that low agreement between field experts on fundamental lexical and orthographic issues, and the lack of a unified editorial standard in available resources make it impossible to build reliable evaluation

This work empirically investigates punctuation insertions as adversarial attacks on NLP systems. Data from experiments on three tasks, five datasets, and six models with four attacks show that punctuation insertions, when limited to a few symbols (apostrophes and hyphens), are a superior attack vector compared to character insertions due to 1) a lower after-attack accuracy ($A_{aft-atk}$) than alphabetical character insertions; 2) higher semantic similarity between the resulting and original texts; and 3) a resulting text that is easier and faster to read as assessed with the Test of Word Reading Efficiency (TOWRE)). The tests also indicate

While multimodal sentiment analysis (MSA) has gained much attention over the last few years, the main focus of most work on MSA has been limited to constructing multimodal representations that capture interactions between different modalities in a single task. This was largely due to a lack of unimodal annotations in MSA benchmark datasets. However, training a model using only multimodal representations can lead to suboptimal performance due to insufficient learning of each uni-modal representation. In this work, to fully optimize learning representations from multimodal data, we propose SUGRM which jointly trains multimodal and unimodal tasks using

Searching troves of videos with textual descriptions is a core multimodal retrieval task. Owing to the lack of a purpose-built dataset for text-to-video retrieval, video captioning datasets have been re-purposed to evaluate models by (1) treating captions as positive matches to their respective videos and (2) assuming all other videos to be negatives. However, this methodology leads to a fundamental flaw during evaluation: since captions are marked as relevant only to their original video, many alternate videos also match the caption, which introduces false-negative caption-video pairs. We show that when these false negatives are corrected, a

Recent advances in text-to-image synthesis make it possible to visualize machine imaginations for a given context. On the other hand, when generating text, human writers are gifted at creative visualization, which enhances their writings by forming imaginations as blueprints before putting down the stories in words. Inspired by such a cognitive process, we ask the natural question of whether we can endow machines with the same ability to utilize visual information and construct a general picture of the context to guide text generation. In this work, we propose iNLG that uses machine-generated images to guide language

We propose a new commonsense reasoning benchmark to motivate commonsense reasoning progress from two perspectives: (1) Evaluating whether models can distinguish knowledge quality by predicting if the knowledge is enough to answer the question; (2) Evaluating whether models can develop commonsense inference capabilities that generalize across tasks. We first extract supporting knowledge for each question and ask humans to annotate whether the auto-extracted knowledge is enough to answer the question or not. After that, we convert different tasks into a unified question-answering format to evaluate the models{'} generalization capabilities. We name the benchmark Commonsense Inference with

Scarcity of large-scale datasets, especially for resource-impoverished languages motivates exploration of data-efficient methods for hate speech detection. Hateful intents are expressed explicitly (use of cuss, swear, abusive words) and implicitly (indirect and contextual). In this work, we progress implicit and explicit hate speech detection using an input-level data augmentation technique, task reformulation using entailment and cross-learning across five languages. Our proposed data augmentation technique EasyMix, improves the performance across all english datasets by {\textasciitilde}1{\%} and across multilingual datasets by {\textasciitilde}1-9{\%}. We also observe substantial gains of {\textasciitilde}2-8{\%} by reformulating hate speech detection as entail problem. We

Large Language Models (LLMs) handle physical commonsense information inadequately. As a result of being trained in a disembodied setting, LLMs often fail to predict an action{'}s outcome in a given environment. However, predicting the effects of an action before it is executed is crucial in planning, where coherent sequences of actions are often needed to achieve a goal. Therefore, we introduce the multi-modal task of predicting the outcomes of actions solely from realistic sensory inputs (images and text). Next, we extend an LLM to model latent representations of objects to better predict action outcomes in an

Leveraging contextual knowledge has become standard practice in automated claim verification, yet the impact of temporal reasoning has been largely overlooked. Our study demonstrates that time positively influences the claim verification process of evidence-based fact-checking. The temporal aspects and relations between claims and evidence are first established through grounding on shared timelines, which are constructed using publication dates and time expressions extracted from their text. Temporal information is then provided to RNN-based and Transformer-based classifiers before or after claim and evidence encoding. Our time-aware fact-checking models surpass base models by up to 9{\%} Micro F1 (64.17{\%})

To mitigate the impact of the scarcity of labelled data on fact-checking systems, we focus on few-shot claim verification. Despite recent work on few-shot classification by proposing advanced language models, there is a dearth of research in data annotation prioritisation that improves the selection of the few shots to be labelled for optimal model performance. We propose Active PETs, a novel weighted approach that utilises an ensemble of Pattern Exploiting Training (PET) models based on various language models, to actively select unlabelled data as candidates for annotation. Using Active PETs for few-shot data selection shows consistent

Table-to-text generation aims at automatically generating text to help people conveniently obtain salient information in tables. Recent works explicitly decompose the generation process into content planning and surface generation stages, employing two autoregressive networks for them respectively.However, they are computationally expensive due to the non-parallelizable nature of autoregressive decoding and the redundant parameters of two networks.In this paper, we propose the first totally non-autoregressive table-to-text model (Plan-then-Seam, PTS) that produces its outputs in parallel with one single network.PTS firstly writes and calibrates one plan of the content to be generated with a novel rethinking pointer predictor,

Prior work in semantic parsing has shown that conventional seq2seq models fail at compositional generalization tasks. This limitation led to a resurgence of methods that model alignments between sentences and their corresponding meaning representations, either implicitly through latent variables or explicitly by taking advantage of alignment annotations. We take the second direction and propose TPol, a two-step approach that first translates input sentences monotonically and then reorders them to obtain the correct output. This is achieved with a modular framework comprising a Translator and a Reorderer component. We test our approach on two popular semantic parsing