We introduce the Dutch Model Benchmark: DUMB. The benchmark includes a diverse set of datasets for low-, medium- and high-resource tasks. The total set of nine tasks includes four tasks that were previously not available in Dutch. Instead of relying on a mean score across tasks, we propose Relative Error Reduction (RER), which compares the DUMB performance of language models to a strong baseline which can be referred to in the future even when assessing different sets of language models. Through a comparison of 14 pre-trained language models (mono- and multi-lingual, of varying sizes), we assess the internal consistency of the benchmark tasks, as well as the factors that likely enable high performance. Our results indicate that current Dutch monolingual models under-perform and suggest training larger Dutch models with other architectures and pre-training objectives. At present, the highest performance is achieved by DeBERTaV3 (large), XLM-R (large) and mDeBERTaV3 (base). In addition to highlighting best strategies for training larger Dutch models, DUMB will foster further research on Dutch. A public leaderboard is available at https://dumbench.nl.

We investigate the usage of auxiliary and modal verbs in Low Saxon dialects from both Germany and the Netherlands based on word vectors, and compare developments in the modern language to Middle Low Saxon. Although most of these function words have not been affected by lexical replacement, changes in usage that likely at least partly result from contact with the state languages can still be observed.

The performance of automatic speech recognition (ASR) systems has advanced substantially in recent years, particularly for languages for which a large amount of transcribed speech is available. Unfortunately, for low-resource languages, such as minority languages, regional languages or dialects, ASR performance generally remains much lower. In this study, we investigate whether data augmentation techniques could help improve low-resource ASR performance, focusing on four typologically diverse minority languages or language variants (West Germanic: Gronings, West-Frisian; Malayo-Polynesian: Besemah, Nasal). For all four languages, we examine the use of self-training, where an ASR system trained with the available human-transcribed data is used to generate transcriptions, which are then combined with the original data to train a new ASR system. For Gronings, for which there was a pre-existing text-to-speech (TTS) system available, we also examined the use of TTS to generate ASR training data from text-only sources. We find that using a self-training approach consistently yields improved performance (a relative WER reduction up to 20.5% compared to using an ASR system trained on 24 minutes of manually transcribed speech). The performance gain from TTS augmentation for Gronings was even stronger (up to 25.5% relative reduction in WER compared to a system based on 24 minutes of manually transcribed speech). In sum, our results show the benefit of using self-training or (if possible) TTS-generated data as an efficient solution to overcome the limitations of data availability for resource-scarce languages in order to improve ASR performance.

Cross-lingual transfer learning with large multilingual pre-trained models can be an effective approach for low-resource languages with no labeled training data. Existing evaluations of zero-shot cross-lingual generalisability of large pre-trained models use datasets with English training data, and test data in a selection of target languages. We explore a more extensive transfer learning setup with 65 different source languages and 105 target languages for part-of-speech tagging. Through our analysis, we show that pre-training of both source and target language, as well as matching language families, writing systems, word order systems, and lexical-phonetic distance significantly impact cross-lingual performance. The findings described in this paper can be used as indicators of which factors are important for effective zero-shot cross-lingual transfer to zero- and low-resource languages.

Deep acoustic models represent linguistic information based on massive amounts of data. Unfortunately, for regional languages and dialects such resources are mostly not available. However, deep acoustic models might have learned linguistic information that transfers to low-resource languages. In this study, we evaluate whether this is the case through the task of distinguishing low-resource (Dutch) regional varieties. By extracting embeddings from the hidden layers of various wav2vec 2.0 models (including new models which are pre-trained and/or fine-tuned on Dutch) and using dynamic time warping, we compute pairwise pronunciation differences averaged over 10 words for over 100 individual dialects from four (regional) languages. We then cluster the resulting difference matrix in four groups and compare these to a gold standard, and a partitioning on the basis of comparing phonetic transcriptions. Our results show that acoustic models outperform the (traditional) transcription-based approach without requiring phonetic transcriptions, with the best performance achieved by the multilingual XLSR-53 model fine-tuned on Dutch. On the basis of only six seconds of speech, the resulting clustering closely matches the gold standard.

We compare five Low Saxon dialects from the 19th and 21st century from Germany and the Netherlands with each other as well as with modern Standard Dutch and Standard German. Our comparison is based on character n-grams on the one hand and PoS n-grams on the other and we show that these two lead to different distances. Particularly in the PoS-based distances, one can observe all of the 21st century Low Saxon dialects shifting towards the modern majority languages.

We present a new comprehensive dataset for the unstandardised West-Germanic language Low Saxon covering the last two centuries, the majority of modern dialects and various genres, which will be made openly available in connection with the final version of this paper. Since so far no such comprehensive dataset of contemporary Low Saxon exists, this provides a great contribution to NLP research on this language. We also test the use of this dataset for dialect classification by training a few baseline models comparing statistical and neural approaches. The performance of these models shows that in spite of an imbalance in the amount of data per dialect, enough features can be learned for a relatively high classification accuracy.

Given (i) the rise of a new paradigm to machine translation based on neural networks that results in more fluent and less literal output than previous models and (ii) the maturity of machine-assisted translation via post-editing in industry, project PiPeNovel studies the feasibility of the post-editing workflow for literary text conducting experiments with professional literary translators.

This study focuses on an essential precondition for reproducibility in computational linguistics: the willingness of authors to share relevant source code and data. Ten years after Ted Pedersen’s influential “Last Words” contribution in Computational Linguistics, we investigate to what extent researchers in computational linguistics are willing and able to share their data and code. We surveyed all 395 full papers presented at the 2011 and 2016 ACL Annual Meetings, and identified whether links to data and code were provided. If working links were not provided, authors were requested to provide this information. Although data were often available, code was shared less often. When working links to code or data were not provided in the paper, authors provided the code in about one third of cases. For a selection of ten papers, we attempted to reproduce the results using the provided data and code. We were able to reproduce the results approximately for six papers. For only a single paper did we obtain the exact same results. Our findings show that even though the situation appears to have improved comparing 2016 to 2011, empiricism in computational linguistics still largely remains a matter of faith. Nevertheless, we are somewhat optimistic about the future. Ensuring reproducibility is not only important for the field as a whole, but also seems worthwhile for individual researchers: The median citation count for studies with working links to the source code is higher.

In this paper, we explore the performance of a linear SVM trained on language independent character features for the NLI Shared Task 2017. Our basic system (GRONINGEN) achieves the best performance (87.56 F1-score) on the evaluation set using only 1-9 character n-grams as features. We compare this against several ensemble and meta-classifiers in order to examine how the linear system fares when combined with other, especially non-linear classifiers. Special emphasis is placed on the topic bias that exists by virtue of the assessment essay prompt distribution.

In this paper, we illustrate the integration of an online dialectometric tool, Gabmap, together with an online dialect atlas, the Atlante Lessicale Toscano (ALT-Web). By using a newly created url-based interface to Gabmap, ALT-Web is able to take advantage of the sophisticated dialect visualization and exploration options incorporated in Gabmap. For example, distribution maps showing the distribution in the Tuscan dialect area of a specific dialectal form (selected via the ALT-Web website) are easily obtainable. Furthermore, the complete ALT-Web dataset as well as subsets of the data (selected via the ALT-Web website) can be automatically uploaded and explored in Gabmap. By combining these two online applications, macro- and micro-analyses of dialectal data (respectively offered by Gabmap and ALT-Web) are effectively and dynamically combined.