Neural machine translation (NMT) systems amplify lexical biases present in their training data, leading to artificially impoverished language in output translations.These language-level characteristics render automatic translations different from text originally written in a language and human translations, which hinders their usefulness in for example creating evaluation datasets.Attempts to increase naturalness in NMT can fall short in terms of content preservation, where increased lexical diversity comes at the cost of translation accuracy.Inspired by the reinforcement learning from human feedback framework, we introduce a novel method that rewards both naturalness and content preservation.We experiment with multiple perspectives to produce more natural translations, aiming at reducing machine and human translationese.We evaluate our method on English-to-Dutch literary translation, and find that our best model produces translations that are lexically richer and exhibit more properties of human-written language, without loss in translation accuracy.

Parallel corpora play a vital role in advanced multilingual natural language processing tasks, notably in machine translation (MT). The recent emergence of numerous large parallel corpora, often extracted from multilingual documents on the Internet, has expanded the available resources. Nevertheless, the quality of these corpora remains largely unexplored, while there are large differences in how the corpora are constructed. Moreover, how the potential differences affect the performance of neural MT (NMT) systems has also received limited attention. This study addresses this gap by manually and automatically evaluating four well-known publicly available parallel corpora across eleven language pairs. Our findings are quite concerning: all corpora contain a substantial amount of noisy sentence pairs, with CCMatrix and CCAligned having well below of 50% reasonably clean pairs. MaCoCu and ParaCrawl generally have higher quality texts, though around a third of the texts still have clear issues. While corpus size impacts NMT models’ performance, our study highlights the critical role of quality: higher-quality corpora consistently yield better-performing NMT models when controlling for size.

Neural text-based models for detecting machine-translated texts can rely on named entities (NEs) as classification shortcuts.While masking NEs encourages learning genuine translationese signals, it degrades the classification performance.Incorporating speech features compensates for this loss, but their interaction with NE reliance requires careful investigation.Through systematic attribution analysis across modalities, we find that bimodal integration leads to more balanced feature utilization, reducing the reliance on NEs in text while moderating overemphasis attribution patterns in speech features.

Discriminating between original texts and machine translations involves identifying whether a text was originally authored in the target language or generated through machine translation. To our knowledge, all methods to date depend exclusively on text-based features. In this study, we move beyond this unimodal approach by incorporating speech features. Machine-translated texts display linguistic deviations from original texts, such as those in lexicon and syntax, which can also manifest in speech characteristics. We evaluate the effectiveness of using text features, speech features, and their bimodal fusion to train classifiers capable of discerning original from machine-translated texts. Additionally, we explore various classification algorithms and fusion techniques. Our results show that speech features alone surpass chance accuracy, while combining text and speech features enhances performance beyond text-only methods. Furthermore, although no single classification or fusion method proves consistently superior, advanced fusion techniques outperform simple feature concatenation.

This study examines the variability of ChatGPT’s machine translation (MT) outputs across six different configurations in four languages, with a focus on creativity in a literary text. We evaluate GPT translations in different text granularity levels, temperature settings and prompting strategies with a Creativity Score formula. We found that prompting ChatGPT with a minimal instruction yields the best creative translations, with Translate the following text into [TG] creatively at the temperature of 1.0 outperforming other configurations and DeepL in Spanish, Dutch, and Chinese. Nonetheless, ChatGPT consistently underperforms compared to human translation (HT). All the code and data are available at Repository URL will be provided with camera-ready version.

Large, curated, web-crawled corpora play a vital role in training language models (LMs). They form the lion’s share of the training data in virtually all recent LMs, such as the well-known GPT, LLaMA and XLM-RoBERTa models. However, despite this importance, relatively little attention has been given to the quality of these corpora. In this paper, we compare four of the currently most relevant large, web-crawled corpora (CC100, MaCoCu, mC4 and OSCAR) across eleven lower-resourced European languages. Our approach is two-fold: first, we perform an intrinsic evaluation by performing a human evaluation of the quality of samples taken from different corpora; then, we assess the practical impact of the qualitative differences by training specific LMs on each of the corpora and evaluating their performance on downstream tasks. We find that there are clear differences in quality of the corpora, with MaCoCu and OSCAR obtaining the best results. However, during the extrinsic evaluation, we actually find that the CC100 corpus achieves the highest scores. We conclude that, in our experiments, the quality of the web-crawled corpora does not seem to play a significant role when training LMs.

The effectiveness of neural machine translation is markedly constrained in low-resource scenarios, where the scarcity of parallel data hampers the development of robust models. This paper focuses on the scenario where the source language is low-resourceand there exists a related high-resource language, for which we introduce a novel approach that combines pivot translation and multilingual training. As a use case we tackle the automatic translation from Catalan to Chinese, using Spanish as an additional language. Our evaluation, conducted on the FLORES-200 benchmark, compares our new approach against a vanilla baseline alongside other models representing various low-resource techniques in the Catalan-to-Chinese context. Experimental results highlight the efficacy of our proposed method, which outperforms existing models, notably demonstrating significant improvements both in translation quality and in lexical diversity.

LT-LiDER is an Erasmus+ cooperation project with two main aims. The first is to map the landscape of technological capabilities required to work as a language and/or translation expert in the digitalised and datafied language industry. The second is to generate training outputs that will help language and translation trainers improve their skills and adopt appropriate pedagogical approaches and strategies for integrating data-driven technology into their language or translation classrooms, with a focus on digital and AI literacy.

Machine translations are found to be lexically poorer than human translations. The loss of lexical diversity through MT poses an issue in the automatic translation of litrature, where it matters not only what is written, but also how it is written. Current methods for increasing lexical diversity in MT are rigid. Yet, as we demonstrate, the degree of lexical diversity can vary considerably across different novels. Thus, rather than aiming for the rigid increase of lexical diversity, we reframe the task as recovering what is lost in the machine translation process. We propose a novel approach that consists of reranking translation candidates with a classifier that distinguishes between original and translated text. We evaluate our approach on 31 English-to-Dutch book translations, and find that, for certain books, our approach retrieves lexical diversity scores that are close to human translation.

Abstract Pretrained character-level and byte-level language models have been shown to be competitive with popular subword models across a range of Natural Language Processing tasks. However, there has been little research on their effectiveness for neural machine translation (NMT), particularly within the popular pretrain-then-finetune paradigm. This work performs an extensive comparison across multiple languages and experimental conditions of character- and subword-level pretrained models (ByT5 and mT5, respectively) on NMT. We show the effectiveness of character-level modeling in translation, particularly in cases where fine-tuning data is limited. In our analysis, we show how character models’ gains in translation quality are reflected in better translations of orthographically similar words and rare words. While evaluating the importance of source texts in driving model predictions, we highlight word-level patterns within ByT5, suggesting an ability to modulate word-level and character-level information during generation. We conclude by assessing the efficiency tradeoff of byte models, suggesting their usage in non-time-critical scenarios to boost translation quality.

We present the most relevant results of the project MaCoCu: Massive collection and curation of monolingual and bilingual data: focus on under-resourced languages in its second year. To date, parallel and monolingual corpora have been produced for seven low-resourced European languages by crawling large amounts of textual data from selected top-level domains of the Internet; both human and automatic evaluation show its usefulness. In addition, several large language models pretrained on MaCoCu data have been published, as well as the code used to collect and curate the data.

We tackle the task of automatically discriminating between human and machine translations. As opposed to most previous work, we perform experiments in a multilingual setting, considering multiple languages and multilingual pretrained language models. We show that a classifier trained on parallel data with a single source language (in our case German–English) can still perform well on English translations that come from different source languages, even when the machine translations were produced by other systems than the one it was trained on. Additionally, we demonstrate that incorporating the source text in the input of a multilingual classifier improves (i) its accuracy and (ii) its robustness on cross-system evaluation, compared to a monolingual classifier. Furthermore, we find that using training data from multiple source languages (German, Russian and Chinese) tends to improve the accuracy of both monolingual and multilingual classifiers. Finally, we show that bilingual classifiers and classifiers trained on multiple source languages benefit from being trained on longer text sequences, rather than on sentences.

We present here the EU-funded project CREAMT that seeks to understand what is meant by creativity in different translation modalities, e.g. machine translation, post-editing or professional translation. Focusing on the textual elements that determine creativity in translated literary texts and the reader experience, CREAMT uses a novel, interdisciplinary approach to assess how effective MT is in literary translation considering creativity in translation and the ultimate user: the reader.

We introduce the project “MaCoCu: Massive collection and curation of monolingual and bilingual data: focus on under-resourced languages”, funded by the Connecting Europe Facility, which is aimed at building monolingual and parallel corpora for under-resourced European languages. The approach followed consists of crawling large amounts of textual data from carefully selected top-level domains of the Internet, and then applying a curation and enrichment pipeline. In addition to corpora, the project will release successive versions of the free/open-source web crawling and curation software used.

We address the task of automatically distinguishing between human-translated (HT) and machine translated (MT) texts. Following recent work, we fine-tune pre-trained language models (LMs) to perform this task. Our work differs in that we use state-of-the-art pre-trained LMs, as well as the test sets of the WMT news shared tasks as training data, to ensure the sentences were not seen during training of the MT system itself. Moreover, we analyse performance for a number of different experimental setups, such as adding translationese data, going beyond the sentence-level and normalizing punctuation. We show that (i) choosing a state-of-the-art LM can make quite a difference: our best baseline system (DeBERTa) outperforms both BERT and RoBERTa by over 3% accuracy, (ii) adding translationese data is only beneficial if there is not much data available, (iii) considerable improvements can be obtained by classifying at the document-level and (iv) normalizing punctuation and thus avoiding (some) shortcuts has no impact on model performance.

We introduce DivEMT, the first publicly available post-editing study of Neural Machine Translation (NMT) over a typologically diverse set of target languages.Using a strictly controlled setup, 18 professional translators were instructed to translate or post-edit the same set of English documents into Arabic, Dutch, Italian, Turkish, Ukrainian, and Vietnamese.During the process, their edits, keystrokes, editing times and pauses were recorded, enabling an in-depth, cross-lingual evaluation of NMT quality and post-editing effectiveness.Using this new dataset, we assess the impact of two state-of-the-art NMT systems, Google Translate and the multilingual mBART-50 model, on translation productivity.We find that post-editing is consistently faster than translation from scratch.However, the magnitude of productivity gains varies widely across systems and languages, highlighting major disparities in post-editing effectiveness for languages at different degrees of typological relatedness to English, even when controlling for system architecture and training data size.We publicly release the complete dataset 1 including all collected behavioral data, to foster new research on the translation capabilities of NMT systems for typologically diverse languages.

Style transfer aims to rewrite a source text in a different target style while preserving its content.We propose a novel approach to this task that leverages generic resources, and without using any task-specific parallel (source-target) data outperforms existing unsupervised approaches on the two most popular style transfer tasks: formality transfer and polarity swap.In practice, we adopt a multistep procedure which builds on a generic pretrained sequence-to-sequence model (BART).First, we strengthen the model's ability to rewrite by further pre-training BART on both an existing collection of generic paraphrases, as well as on synthetic pairs created using a general-purpose lexical resource.Second, through an iterative back-translation approach, we train two models, each in a transfer direction, so that they can provide each other with synthetically generated pairs, dynamically in the training process.Lastly, we let our best resulting model generate static synthetic pairs to be used in a supervised training regime.Besides methodology and state-of-the-art results, a core contribution of this work is a reflection on the nature of the two tasks we address, and how their differences are highlighted by their response to our approach.

Unsupervised Machine Translation has been advancing our ability to translate without parallel data, but state-of-the-art methods assume an abundance of monolingual data. This paper investigates the scenario where monolingual data is limited as well, finding that current unsupervised methods suffer in performance under this stricter setting. We find that the performance loss originates from the poor quality of the pretrained monolingual embeddings, and we offer a potential solution: dependency-based word embeddings. These embeddings result in a complementary word representation which offers a boost in performance of around 1.5 BLEU points compared to standard word2vec when monolingual data is limited to 1 million sentences per language. We also find that the inclusion of sub-word information is crucial to improving the quality of the embeddings.

We reassess the claims of human parity and super-human performance made at the news shared task of WMT2019 for three translation directions: English→German, English→Russian and German→English. First we identify three potential issues in the human evaluation of that shared task: (i) the limited amount of intersen- tential context available, (ii) the limited translation proficiency of the evaluators and (iii) the use of a reference transla- tion. We then conduct a modified eval- uation taking these issues into account. Our results indicate that all the claims of human parity and super-human perfor- mance made at WMT2019 should be re- futed, except the claim of human parity for English→German. Based on our findings, we put forward a set of recommendations and open questions for future assessments of human parity in machine translation.

This research presents a fine-grained human evaluation to compare the Transformer and recurrent approaches to neural machine translation (MT), on the translation direction English-to-Chinese. To this end, we develop an error taxonomy compliant with the Multidimensional Quality Metrics (MQM) framework that is customised to the relevant phenomena of this translation direction. We then conduct an error annotation using this customised error taxonomy on the output of state-of-the-art recurrent- and Transformer-based MT systems on a subset of WMT2019’s news test set. The resulting annotation shows that, compared to the best recurrent system, the best Transformer system results in a 31% reduction of the total number of errors and it produced significantly less errors in 10 out of 22 error categories. We also note that two of the systems evaluated do not produce any error for a category that was relevant for this translation direction prior to the advent of NMT systems: Chinese classifiers.

We combine character-level and contextual language model representations to improve performance on Discourse Representation Structure parsing.Character representations can easily be added in a sequence-to-sequence model in either one encoder or as a fully separate encoder, with improvements that are robust to different language models, languages and data sets.For English, these improvements are larger than adding individual sources of linguistic information or adding non-contextual embeddings.A new method of analysis based on semantic tags demonstrates that the character-level representations improve performance across a subset of selected semantic phenomena.

The quality of machine translation has increased remarkably over the past years, to the degree that it was found to be indistinguishable from professional human translation in a number of empirical investigations. We reassess Hassan et al.'s 2018 investigation into Chinese to English news translation, showing that the finding of human–machine parity was owed to weaknesses in the evaluation design—which is currently considered best practice in the field. We show that the professional human translations contained significantly fewer errors, and that perceived quality in human evaluation depends on the choice of raters, the availability of linguistic context, and the creation of reference translations. Our results call for revisiting current best practices to assess strong machine translation systems in general and human–machine parity in particular, for which we offer a set of recommendations based on our empirical findings.

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.

Neural methods have had several recent successes in semantic parsing, though they have yet to face the challenge of producing meaning representations based on formal semantics. We present a sequence-to-sequence neural semantic parser that is able to produce Discourse Representation Structures (DRSs) for English sentences with high accuracy, outperforming traditional DRS parsers. To facilitate the learning of the output, we represent DRSs as a sequence of flat clauses and introduce a method to verify that produced DRSs are well-formed and interpretable. We compare models using characters and words as input and see (somewhat surprisingly) that the former performs better than the latter. We show that eliminating variable names from the output using De Bruijn indices increases parser performance. Adding silver training data boosts performance even further.

We aim to shed light on the strengths and weaknesses of the newly introduced neural machine translation paradigm.To that end, we conduct a multifaceted evaluation in which we compare outputs produced by state-of-the-art neural machine translation and phrase-based machine translation systems for 9 language directions across a number of dimensions.Specifically, we measure the similarity of the outputs, their fluency and amount of reordering, the effect of sentence length and performance across different error categories.We find out that translations produced by neural machine translation systems are considerably different, more fluent and more accurate in terms of word order compared to those produced by phrase-based systems.Neural machine translation systems are also more accurate at producing inflected forms, but they perform poorly when translating very long sentences.

We present a syntax-based reordering model (RM) for hierarchical phrase-based statistical machine translation (HPB-SMT) enriched with semantic features. Our model brings a number of novel contributions: (i) while the previous dependency-based RM is limited to the reordering of head and dependant constituent pairs, we also model the reordering of pairs of dependants; (ii) Our model is enriched with semantic features (Wordnet synsets) in order to allow the reordering model to generalize to pairs not seen in training but with equivalent meaning. (iii) We evaluate our model on two language directions: English-to-Farsi and English-to-Turkish. These language pairs are particularly challenging due to the free word order, rich morphology and lack of resources of the target languages. We evaluate our RM both intrinsically (accuracy of the RM classifier) and extrinsically (MT). Our best configuration outperforms the baseline classifier by 5–29% on pairs of dependants and by 12–30% on head and dependant pairs while the improvement on MT ranges between 1.6% and 5.5% relative in terms of BLEU depending on language pair and domain. We also analyze the value of the feature weights to obtain further insights on the impact of the reordering-related features in the HPB-SMT model. We observe that the features of our RM are assigned significant weights and that our features are complementary to the reordering feature included by default in the HPB-SMT model.

We propose the use of WordNet synsets in a syntax-based reordering model for hierarchical statistical machine translation (HPB-SMT) to enable the model to generalize to phrases not seen in the training data but that have equivalent meaning.We detail our methodology to incorporate synsets' knowledge in the reordering model and evaluate the resulting WordNetenhanced SMT systems on the English-to-Farsi language direction.The inclusion of synsets leads to the best BLEU score, outperforming the baseline (standard HPB-SMT) by 0.6 points absolute.

This paper explores the use of linguistic information for the selection of data to train language models. We depart from the state-of-the-art method in perplexity-based data selection and extend it in order to use word-level linguistic units (i.e. lemmas, named entity categories and part-of-speech tags) instead of surface forms. We then present two methods that combine the different types of linguistic knowledge as well as the surface forms (1, naïve selection of the top ranked sentences selected by each method; 2, linear interpolation of the datasets selected by the different methods). The paper presents detailed results and analysis for four languages with different levels of morphologic complexity (English, Spanish, Czech and Chinese). The interpolation-based combination outperforms the purely statistical baseline in all the scenarios, resulting in language models with lower perplexity. In relative terms the improvements are similar regardless of the language, with perplexity reductions achieved in the range 7.72–13.02%. In absolute terms the reduction is higher for languages with high type-token ratio (Chinese, 202.16) or rich morphology (Czech, 81.53) and lower for the remaining languages, Spanish (55.2) and English (34.43 on the English side of the same parallel dataset as for Czech and 61.90 on the same parallel dataset as for Spanish).

Aswarth Abhilash Dara, Josef van Genabith, Qun Liu, John Judge, Antonio Toral. Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics, volume 2: Short Papers. 2014.

We develop an open-source large-scale finite-state morphological processing toolkit (AraComLex) for Modern Standard Arabic (MSA) distributed under the GPLv3 license (http://aracomlex.sourceforge.net). The morphological transducer is based on a lexical database specifically constructed for this purpose. In contrast to previous resources, the database is tuned to MSA, eliminating lexical entries no longer attested in contemporary use. The database is built using a corpus of 1,089,111,204 word tokens, a pre-annotation tool, machine learning techniques and knowledge-based pattern matching to automatically acquire lexical knowledge. Our morphological transducer is evaluated and compared to LDC's SAMA(Standard Arabic Morphological Analyser). We also develop a finite-state morphological guesser as part of a methodology for extracting unknown word forms, lemmatizing them, and giving them a priority weight for inclusion in the lexicon.