We present Belebele, a multiple-choice machine reading comprehension (MRC) dataset spanning 122 language variants. Significantly expanding the language coverage of natural language understanding (NLU) benchmarks, this dataset enables the evaluation of text models in high-, medium-, and low-resource languages. Each question is based on a short passage from the FLORES-200 dataset and has four multiple-choice answers. The questions were carefully curated to discriminate between models with different levels of general language comprehension. The English dataset on its own proves difficult enough to challenge state-of-the-art language models. Being fully parallel, this dataset enables direct comparison of model performance across all languages. We use this dataset to evaluate the capabilities of multilingual masked language models (MLMs) and large language models (LLMs). We present extensive results and findings, notably that despite significant cross-lingual transfer in English-centric LLMs, much smaller MLMs pretrained on balanced multilingual data still understand far more languages. Overall, Belebele opens up new avenues for evaluating and analyzing the multilingual capabilities of NLP systems.

Multilingual generative language models (LMs) are increasingly fluent in a large variety of languages. Trained on the concatenation of corpora in multiple languages, they enable powerful transfer from high-resource languages to low-resource ones. However, it is still unknown what cultural biases are induced in the predictions of these models. In this work, we focus on one language property highly influenced by culture: formality. We analyze the formality distributions of XGLM and BLOOM’s predictions, two popular generative multilingual language models, in 5 languages. We classify 1,200 generations per language as formal, informal, or incohesive and measure the impact of the prompt formality on the predictions. Overall, we observe a diversity of behaviors across the models and languages. For instance, XGLM generates informal text in Arabic and Bengali when conditioned with informal prompts, much more than BLOOM. In addition, even though both models are highly biased toward the formal style when prompted neutrally, we find that the models generate a significant amount of informal predictions even when prompted with formal text. We release with this work 6,000 annotated samples, paving the way for future work on the formality of generative multilingual LMs.

Trustworthy answer content is abundant in many high-resource languages and is instantly accessible through question answering systems — yet this content can be hard to access for those that do not speak these languages. The leap forward in cross-lingual modeling quality offered by generative language models offers much promise, yet their raw generations often fall short in factuality. To improve trustworthiness in these systems, a promising direction is to attribute the answer to a retrieved source, possibly in a content-rich language different from the query. Our work is the first to study attribution for cross-lingual question answering. First, we collect data in 5 languages to assess the attribution level of a state-of-the-art cross-lingual QA system. To our surprise, we find that a substantial portion of the answers is not attributable to any retrieved passages (up to 50% of answers exactly matching a gold reference) despite the system being able to attend directly to the retrieved text. Second, to address this poor attribution level, we experiment with a wide range of attribution detection techniques. We find that Natural Language Inference models and PaLM 2 fine-tuned on a very small amount of attribution data can accurately detect attribution. With these models, we improve the attribution level of a cross-lingual QA system. Overall, we show that current academic generative cross-lingual QA systems have substantial shortcomings in attribution and we build tooling to mitigate these issues.

Gender biases in language generation systems are challenging to mitigate. One possible source for these biases is gender representation disparities in the training and evaluation data. Despite recent progress in documenting this problem and many attempts at mitigating it, we still lack shared methodology and tooling to report gender representation in large datasets. Such quantitative reporting will enable further mitigation, e.g., via data augmentation. This paper describes the Gender-Gap Pipeline (for Gender-Aware Polyglot Pipeline), an automatic pipeline to characterize gender representation in large-scale datasets for 55 languages. The pipeline uses a multilingual lexicon of gendered person-nouns to quantify the gender representation in text. We showcase it to report gender representation in WMT training data and development data for the News task, confirming that current data is skewed towards masculine representation. Having unbalanced datasets may indirectly optimize our systems towards outperforming one gender over the others. We suggest introducing our gender quantification pipeline in current datasets and, ideally, modifying them toward a balanced representation.

The 2022 Multilingual Representation Learning (MRL) Shared Task was dedicated to clause-level morphology. As the first ever benchmark that defines and evaluates morphology outside its traditional lexical boundaries, the shared task on multilingual clause-level morphology sets the scene for competition across different approaches to morphological modeling, with 3 clause-level sub-tasks: morphological inflection, reinflection and analysis, where systems are required to generate, manipulate or analyze simple sentences centered around a single content lexeme and a set of morphological features characterizing its syntactic clause. This year’s tasks covered eight typologically distinct languages: English, French, German, Hebrew, Russian, Spanish, Swahili and Turkish. The tasks has received submissions of four systems from three teams which were compared to two baselines implementing prominent multilingual learning methods. The results show that modern NLP models are effective in solving morphological tasks even at the clause level. However, there is still room for improvement, especially in the task of morphological analysis.

Open-Domain Generative Question Answering has achieved impressive performance in English by combining document-level retrieval with answer generation. These approaches, which we refer to as GenQA, can generate complete sentences, effectively answering both factoid and non-factoid questions. In this paper, we extend to the multilingual and cross-lingual settings. For this purpose, we first introduce GenTyDiQA, an extension of the TyDiQA dataset with well-formed and complete answers for Arabic, Bengali, English, Japanese, and Russian. Based on GenTyDiQA, we design a cross-lingual generative model that produces full-sentence answers by exploiting passages written in multiple languages, including languages different from the question. Our cross-lingual generative system outperforms answer sentence selection baselines for all 5 languages and monolingual generative pipelines for three out of five languages studied.

L’apprentissage par transfert basé sur le pré-entraînement de modèles de langue sur une grande quantité de données brutes est devenu la norme pour obtenir des performances état de l’art en TAL. Cependant, la façon dont cette approche devrait être appliquée pour des langues inconnues, qui ne sont couvertes par aucun modèle de langue multilingue à grande échelle et pour lesquelles seule une petite quantité de données brutes est le plus souvent disponible, n’est pas claire. Dans ce travail, en comparant des modèles multilingues et monolingues, nous montrons que de tels modèles se comportent de multiples façons sur des langues inconnues. Certaines langues bénéficient grandement de l’apprentissage par transfert et se comportent de manière similaire à des langues proches riches en ressource, alors que ce n’est manifestement pas le cas pour d’autres. En nous concentrant sur ces dernières, nous montrons dans ce travail que cet échec du transfert est largement lié à l’impact du script que ces langues utilisent. Nous montrons que la translittération de ces langues améliore considérablement le potentiel des larges modèles de langue neuronaux multilingues pour des tâches en aval. Ce résultat indique une piste prometteuse pour rendre ces modèles massivement multilingues utiles pour de nouveaux ensembles de langues absentes des données d’entraînement.

This paper describes the Inria ALMAnaCH team submission to the WMT 2022 general translation shared task. Participating in the language directions cs,ru,uk→en and cs↔uk, we experiment with the use of a dedicated Latin-script transcription convention aimed at representing all Slavic languages involved in a way that maximises character- and word-level correspondences between them as well as with the English language. Our hypothesis was that bringing the source and target language closer could have a positive impact on machine translation results. We provide multiple comparisons, including bilingual and multilingual baselines, with and without transcription. Initial results indicate that the transcription strategy was not successful, resulting in lower results than baselines. We nevertheless submitted our multilingual, transcribed models as our primary systems, and in this paper provide some indications as to why we got these negative results.

Lexical normalization is the task of transforming an utterance into its standardized form. This task is beneficial for downstream analysis, as it provides a way to harmonize (often spontaneous) linguistic variation. Such variation is typical for social media on which information is shared in a multitude of ways, including diverse languages and code-switching. Since the seminal work of Han and Baldwin (2011) a decade ago, lexical normalization has attracted attention in English and multiple other languages. However, there exists a lack of a common benchmark for comparison of systems across languages with a homogeneous data and evaluation setup. The MultiLexNorm shared task sets out to fill this gap. We provide the largest publicly available multilingual lexical normalization benchmark including 13 language variants. We propose a homogenized evaluation setup with both intrinsic and extrinsic evaluation. As extrinsic evaluation, we use dependency parsing and part-of-speech tagging with adapted evaluation metrics (a-LAS, a-UAS, and a-POS) to account for alignment discrepancies. The shared task hosted at W-NUT 2021 attracted 9 participants and 18 submissions. The results show that neural normalization systems outperform the previous state-of-the-art system by a large margin. Downstream parsing and part-of-speech tagging performance is positively affected but to varying degrees, with improvements of up to 1.72 a-LAS, 0.85 a-UAS, and 1.54 a-POS for the winning system.

Transfer learning based on pretraining language models on a large amount of raw data has become a new norm to reach state-of-the-art performance in NLP. Still, it remains unclear how this approach should be applied for unseen languages that are not covered by any available large-scale multilingual language model and for which only a small amount of raw data is generally available. In this work, by comparing multilingual and monolingual models, we show that such models behave in multiple ways on unseen languages. Some languages greatly benefit from transfer learning and behave similarly to closely related high resource languages whereas others apparently do not. Focusing on the latter, we show that this failure to transfer is largely related to the impact of the script used to write such languages. We show that transliterating those languages significantly improves the potential of large-scale multilingual language models on downstream tasks. This result provides a promising direction towards making these massively multilingual models useful for a new set of unseen languages.

Multilingual pretrained language models have demonstrated remarkable zero-shot cross-lingual transfer capabilities. Such transfer emerges by fine-tuning on a task of interest in one language and evaluating on a distinct language, not seen during the fine-tuning. Despite promising results, we still lack a proper understanding of the source of this transfer. Using a novel layer ablation technique and analyses of the model’s internal representations, we show that multilingual BERT, a popular multilingual language model, can be viewed as the stacking of two sub-networks: a multilingual encoder followed by a task-specific language-agnostic predictor. While the encoder is crucial for cross-lingual transfer and remains mostly unchanged during fine-tuning, the task predictor has little importance on the transfer and can be reinitialized during fine-tuning. We present extensive experiments with three distinct tasks, seventeen typologically diverse languages and multiple domains to support our hypothesis.

The French TreeBank developed at the University Paris 7 is the main source of morphosyntactic and syntactic annotations for French. However, it does not include explicit information related to named entities, which are among the most useful information for several natural language processing tasks and applications. Moreover, no large-scale French corpus with named entity annotations contain referential information, which complement the type and the span of each mention with an indication of the entity it refers to. We have manually annotated the French TreeBank with such information, after an automatic pre-annotation step. We sketch the underlying annotation guidelines and we provide a few figures about the resulting annotations.

We introduce the first treebank for a romanized user-generated content variety of Algerian, a North-African Arabic dialect known for its frequent usage of code-switching. Made of 1500 sentences, fully annotated in morpho-syntax and Universal Dependency syntax, with full translation at both the word and the sentence levels, this treebank is made freely available. It is supplemented with 50k unlabeled sentences collected from Common Crawl and web-crawled data using intensive data-mining techniques. Preliminary experiments demonstrate its usefulness for POS tagging and dependency parsing. We believe that what we present in this paper is useful beyond the low-resource language community. This is the first time that enough unlabeled and annotated data is provided for an emerging user-generated content dialectal language with rich morphology and code switching, making it an challenging test-bed for most recent NLP approaches.

Pretrained language models are now ubiquitous in Natural Language Processing. Despite their success, most available models have either been trained on English data or on the concatenation of data in multiple languages. This makes practical use of such models –in all languages except English– very limited. In this paper, we investigate the feasibility of training monolingual Transformer-based language models for other languages, taking French as an example and evaluating our language models on part-of-speech tagging, dependency parsing, named entity recognition and natural language inference tasks. We show that the use of web crawled data is preferable to the use of Wikipedia data. More surprisingly, we show that a relatively small web crawled dataset (4GB) leads to results that are as good as those obtained using larger datasets (130+GB). Our best performing model CamemBERT reaches or improves the state of the art in all four downstream tasks.

Les modèles de langue neuronaux contextuels sont désormais omniprésents en traitement automatique des langues. Jusqu’à récemment, la plupart des modèles disponibles ont été entraînés soit sur des données en anglais, soit sur la concaténation de données dans plusieurs langues. L’utilisation pratique de ces modèles — dans toutes les langues sauf l’anglais — était donc limitée. La sortie récente de plusieurs modèles monolingues fondés sur BERT (Devlin et al., 2019), notamment pour le français, a démontré l’intérêt de ces modèles en améliorant l’état de l’art pour toutes les tâches évaluées. Dans cet article, à partir d’expériences menées sur CamemBERT (Martin et al., 2019), nous montrons que l’utilisation de données à haute variabilité est préférable à des données plus uniformes. De façon plus surprenante, nous montrons que l’utilisation d’un ensemble relativement petit de données issues du web (4Go) donne des résultats aussi bons que ceux obtenus à partir d’ensembles de données plus grands de deux ordres de grandeurs (138Go).

Language model-based pre-trained representations have become ubiquitous in natural language processing. They have been shown to significantly improve the performance of neural models on a great variety of tasks. However, it remains unclear how useful those general models can be in handling non-canonical text. In this article, focusing on User Generated Content (UGC), we study the ability of BERT to perform lexical normalisation. Our contribution is simple: by framing lexical normalisation as a token prediction task, by enhancing its architecture and by carefully fine-tuning it, we show that BERT can be a competitive lexical normalisation model without the need of any UGC resources aside from 3,000 training sentences. To the best of our knowledge, it is the first work done in adapting and analysing the ability of this model to handle noisy UGC data.

In this paper, we present the details of the neural dependency parser and the neural tagger submitted by our team ‘ParisNLP’ to the CoNLL 2018 Shared Task on parsing from raw text to Universal Dependencies. We augment the deep Biaffine (BiAF) parser (Dozat and Manning, 2016) with novel features to perform competitively: we utilize an indomain version of ELMo features (Peters et al., 2018) which provide context-dependent word representations; we utilize disambiguated, embedded, morphosyntactic features from lexicons (Sagot, 2018), which complements the existing feature set. Henceforth, we call our system ‘ELMoLex’. In addition to incorporating character embeddings, ELMoLex benefits from pre-trained word vectors, ELMo and morphosyntactic features (whenever available) to correctly handle rare or unknown words which are prevalent in languages with complex morphology. ELMoLex ranked 11th by Labeled Attachment Score metric (70.64%), Morphology-aware LAS metric (55.74%) and ranked 9th by Bilexical dependency metric (60.70%).