Autoregressive language models are trained by minimizing the cross-entropy of the model distribution Q relative to the data distribution P – that is, minimizing the forward cross-entropy, which is equivalent to maximum likelihood estimation (MLE). We have observed that models trained in this way may “over-generalize”, in the sense that they produce non-human-like text. Moreover, we believe that reverse cross-entropy, i.e., the cross-entropy of P relative to Q, is a better reflection of how a human would evaluate text generated by a model. Hence, we propose learning with MixCE, an objective that mixes the forward and reverse cross-entropies. We evaluate models trained with this objective on synthetic data settings (where P is known) and real data, and show that the resulting models yield better generated text without complex decoding strategies.

Named Entity Recognition is a key Natural Language Processing task whose performance is sensitive to choice of genre and language. A unified NER model across multiple genres and languages is more practical and efficient by leveraging commonalities across genres or languages. In this paper, we propose a novel setup for NER which includes multi-domain and multilingual training and evaluation across 13 domains and 4 languages. We explore a range of approaches to building a unified model using domain and language adaptation techniques. Our experiments highlight multiple nuances to consider while building a unified model, including that naive data pooling fails to obtain good performance, that domain-specific adaptations are more important than language-specific ones and that including domain-specific adaptations in a unified model nears the performance of training multiple dedicated monolingual models at a fraction of their parameter count.

Real-life multilingual systems should be able to efficiently incorporate new languages as data distributions fed to the system evolve and shift over time. To do this, systems need to handle the issue of catastrophic forgetting, where the model performance drops for languages or tasks seen further in its past. In this paper, we study catastrophic forgetting, as well as methods to minimize this, in a massively multilingual continual learning framework involving up to 51 languages and covering both classification and sequence labeling tasks. We present LR ADJUST, a learning rate scheduling method that is simple, yet effective in preserving new information without strongly overwriting past knowledge. Furthermore, we show that this method is effective across multiple continual learning approaches. Finally, we provide further insights into the dynamics of catastrophic forgetting in this massively multilingual setup.

Pretrained multilingual encoders enable zero-shot cross-lingual transfer, but often produce unreliable models that exhibit high performance variance on the target language. We postulate that this high variance results from zero-shot cross-lingual transfer solving an under-specified optimization problem. We show that any linear-interpolated model between the source language monolingual model and source + target bilingual model has equally low source language generalization error, yet the target language generalization error reduces smoothly and linearly as we move from the monolingual to bilingual model, suggesting that the model struggles to identify good solutions for both source and target languages using the source language alone. Additionally, we show that zero-shot solution lies in non-flat region of target language error generalization surface, causing the high variance.

The language of Twitter differs significantly from that of other domains commonly included in large language model training. While tweets are typically multilingual and contain informal language, including emoji and hashtags, most pre-trained language models for Twitter are either monolingual, adapted from other domains rather than trained exclusively on Twitter, or are trained on a limited amount of in-domain Twitter data.We introduce Bernice, the first multilingual RoBERTa language model trained from scratch on 2.5 billion tweets with a custom tweet-focused tokenizer. We evaluate on a variety of monolingual and multilingual Twitter benchmarks, finding that our model consistently exceeds or matches the performance of a variety of models adapted to social media data as well as strong multilingual baselines, despite being trained on less data overall.We posit that it is more efficient compute- and data-wise to train completely on in-domain data with a specialized domain-specific tokenizer.

Large pre-trained language models (LMs) have demonstrated the ability to obtain good performance on downstream tasks with limited examples in cross-lingual settings. However, this was mostly studied for relatively resource-rich languages, where at least enough unlabeled data is available to be included in pre-training a multilingual language model. In this paper, we explore the problem of cross-lingual transfer in unseen languages, where no unlabeled data is available for pre-training a model. We use a downstream sentiment analysis task across 12 languages, including 8 unseen languages, to analyze the effectiveness of several few-shot learning strategies across the three major types of model architectures and their learning dynamics. We also compare strategies for selecting languages for transfer and contrast findings across languages seen in pre-training compared to those that are not. Our findings contribute to the body of knowledge on cross-lingual models for low-resource settings that is paramount to increasing coverage, diversity, and equity in access to NLP technology. We show that, in few-shot learning, linguistically similar and geographically similar languages are useful for cross-lingual adaptation, but taking the context from a mixture of random source languages is surprisingly more effective. We also compare different model architectures and show that the encoder-only model, XLM-R, gives the best downstream task performance.

The transformer has been shown to outperform recurrent neural network-based sequence-to-sequence models in various word-level NLP tasks. Yet for character-level transduction tasks, e.g. morphological inflection generation and historical text normalization, there are few works that outperform recurrent models using the transformer. In an empirical study, we uncover that, in contrast to recurrent sequence-to-sequence models, the batch size plays a crucial role in the performance of the transformer on character-level tasks, and we show that with a large enough batch size, the transformer does indeed outperform recurrent models. We also introduce a simple technique to handle feature-guided character-level transduction that further improves performance. With these insights, we achieve state-of-the-art performance on morphological inflection and historical text normalization. We also show that the transformer outperforms a strong baseline on two other character-level transduction tasks: grapheme-to-phoneme conversion and transliteration.

Zero-shot cross-lingual information extraction (IE) describes the construction of an IE model for some target language, given existing annotations exclusively in some other language, typically English. While the advance of pretrained multilingual encoders suggests an easy optimism of “train on English, run on any language”, we find through a thorough exploration and extension of techniques that a combination of approaches, both new and old, leads to better performance than any one cross-lingual strategy in particular. We explore techniques including data projection and self-training, and how different pretrained encoders impact them. We use English-to-Arabic IE as our initial example, demonstrating strong performance in this setting for event extraction, named entity recognition, part-of-speech tagging, and dependency parsing. We then apply data projection and self-training to three tasks across eight target languages. Because no single set of techniques performs the best across all tasks, we encourage practitioners to explore various configurations of the techniques described in this work when seeking to improve on zero-shot training.

This year’s iteration of the SIGMORPHON Shared Task on morphological reinflection focuses on typological diversity and cross-lingual variation of morphosyntactic features. In terms of the task, we enrich UniMorph with new data for 32 languages from 13 language families, with most of them being under-resourced: Kunwinjku, Classical Syriac, Arabic (Modern Standard, Egyptian, Gulf), Hebrew, Amharic, Aymara, Magahi, Braj, Kurdish (Central, Northern, Southern), Polish, Karelian, Livvi, Ludic, Veps, Võro, Evenki, Xibe, Tuvan, Sakha, Turkish, Indonesian, Kodi, Seneca, Asháninka, Yanesha, Chukchi, Itelmen, Eibela. We evaluate six systems on the new data and conduct an extensive error analysis of the systems’ predictions. Transformer-based models generally demonstrate superior performance on the majority of languages, achieving >90% accuracy on 65% of them. The languages on which systems yielded low accuracy are mainly under-resourced, with a limited amount of data. Most errors made by the systems are due to allomorphy, honorificity, and form variation. In addition, we observe that systems especially struggle to inflect multiword lemmas. The systems also produce misspelled forms or end up in repetitive loops (e.g., RNN-based models). Finally, we report a large drop in systems’ performance on previously unseen lemmas.

A broad goal in natural language processing (NLP) is to develop a system that has the capacity to process any natural language. Most systems, however, are developed using data from just one language such as English. The SIGMORPHON 2020 shared task on morphological reinflection aims to investigate systems’ ability to generalize across typologically distinct languages, many of which are low resource. Systems were developed using data from 45 languages and just 5 language families, fine-tuned with data from an additional 45 languages and 10 language families (13 in total), and evaluated on all 90 languages. A total of 22 systems (19 neural) from 10 teams were submitted to the task. All four winning systems were neural (two monolingual transformers and two massively multilingual RNN-based models with gated attention). Most teams demonstrate utility of data hallucination and augmentation, ensembles, and multilingual training for low-resource languages. Non-neural learners and manually designed grammars showed competitive and even superior performance on some languages (such as Ingrian, Tajik, Tagalog, Zarma, Lingala), especially with very limited data. Some language families (Afro-Asiatic, Niger-Congo, Turkic) were relatively easy for most systems and achieved over 90% mean accuracy while others were more challenging.

We describe the design and findings of the SIGMORPHON 2020 shared task on multilingual grapheme-to-phoneme conversion. Participants were asked to submit systems which take in a sequence of graphemes in a given language as input, then output a sequence of phonemes representing the pronunciation of that grapheme sequence. Nine teams submitted a total of 23 systems, at best achieving a 18% relative reduction in word error rate (macro-averaged over languages), versus strong neural sequence-to-sequence baselines. To facilitate error analysis, we publicly release the complete outputs for all systems—a first for the SIGMORPHON workshop.

We study the problem of multilingual masked language modeling, i.e. the training of a single model on concatenated text from multiple languages, and present a detailed study of several factors that influence why these models are so effective for cross-lingual transfer. We show, contrary to what was previously hypothesized, that transfer is possible even when there is no shared vocabulary across the monolingual corpora and also when the text comes from very different domains. The only requirement is that there are some shared parameters in the top layers of the multi-lingual encoder. To better understand this result, we also show that representations from monolingual BERT models in different languages can be aligned post-hoc quite effectively, strongly suggesting that, much like for non-contextual word embeddings, there are universal latent symmetries in the learned embedding spaces. For multilingual masked language modeling, these symmetries are automatically discovered and aligned during the joint training process.

This work treats the paradigm discovery problem (PDP), the task of learning an inflectional morphological system from unannotated sentences. We formalize the PDP and develop evaluation metrics for judging systems. Using currently available resources, we construct datasets for the task. We also devise a heuristic benchmark for the PDP and report empirical results on five diverse languages. Our benchmark system first makes use of word embeddings and string similarity to cluster forms by cell and by paradigm. Then, we bootstrap a neural transducer on top of the clustered data to predict words to realize the empty paradigm slots. An error analysis of our system suggests clustering by cell across different inflection classes is the most pressing challenge for future work.

Multilingual BERT (mBERT), XLM-RoBERTa (XLMR) and other unsupervised multilingual encoders can effectively learn cross-lingual representation. Explicit alignment objectives based on bitexts like Europarl or MultiUN have been shown to further improve these representations. However, word-level alignments are often suboptimal and such bitexts are unavailable for many languages. In this paper, we propose a new contrastive alignment objective that can better utilize such signal, and examine whether these previous alignment methods can be adapted to noisier sources of aligned data: a randomly sampled 1 million pair subset of the OPUS collection. Additionally, rather than report results on a single dataset with a single model run, we report the mean and standard derivation of multiple runs with different seeds, on four datasets and tasks. Our more extensive analysis finds that, while our new objective outperforms previous work, overall these methods do not improve performance with a more robust evaluation framework. Furthermore, the gains from using a better underlying model eclipse any benefits from alignment training. These negative results dictate more care in evaluating these methods and suggest limitations in applying explicit alignment objectives.

Pretrained contextualized text encoders are now a staple of the NLP community. We present a survey on language representation learning with the aim of consolidating a series of shared lessons learned across a variety of recent efforts. While significant advancements continue at a rapid pace, we find that enough has now been discovered, in different directions, that we can begin to organize advances according to common themes. Through this organization, we highlight important considerations when interpreting recent contributions and choosing which model to use.

Multilingual BERT (mBERT) trained on 104 languages has shown surprisingly good cross-lingual performance on several NLP tasks, even without explicit cross-lingual signals. However, these evaluations have focused on cross-lingual transfer with high-resource languages, covering only a third of the languages covered by mBERT. We explore how mBERT performs on a much wider set of languages, focusing on the quality of representation for low-resource languages, measured by within-language performance. We consider three tasks: Named Entity Recognition (99 languages), Part-of-speech Tagging and Dependency Parsing (54 languages each). mBERT does better than or comparable to baselines on high resource languages but does much worse for low resource languages. Furthermore, monolingual BERT models for these languages do even worse. Paired with similar languages, the performance gap between monolingual BERT and mBERT can be narrowed. We find that better models for low resource languages require more efficient pretraining techniques or more data.

Pretrained contextual representation models (Peters et al., 2018; Devlin et al., 2018) have pushed forward the state-of-the-art on many NLP tasks. A new release of BERT (Devlin, 2018) includes a model simultaneously pretrained on 104 languages with impressive performance for zero-shot cross-lingual transfer on a natural language inference task. This paper explores the broader cross-lingual potential of mBERT (multilingual) as a zero shot language transfer model on 5 NLP tasks covering a total of 39 languages from various language families: NLI, document classification, NER, POS tagging, and dependency parsing. We compare mBERT with the best-published methods for zero-shot cross-lingual transfer and find mBERT competitive on each task. Additionally, we investigate the most effective strategy for utilizing mBERT in this manner, determine to what extent mBERT generalizes away from language specific features, and measure factors that influence cross-lingual transfer.

English verbs have multiple forms. For instance, talk may also appear as talks, talked or talking, depending on the context. The NLP task of lemmatization seeks to map these diverse forms back to a canonical one, known as the lemma. We present a simple joint neural model for lemmatization and morphological tagging that achieves state-of-the-art results on 20 languages from the Universal Dependencies corpora. Our paper describes the model in addition to training and decoding procedures. Error analysis indicates that joint morphological tagging and lemmatization is especially helpful in low-resource lemmatization and languages that display a larger degree of morphological complexity.

Many common character-level, string-to-string transduction tasks, e.g., grapheme-to-phoneme conversion and morphological inflection, consist almost exclusively of monotonic transduction. Neural sequence-to-sequence models with soft attention, non-monotonic models, outperform popular monotonic models. In this work, we ask the following question: Is monotonicity really a helpful inductive bias in these tasks? We develop a hard attention sequence-to-sequence model that enforces strict monotonicity and learns alignment jointly. With the help of dynamic programming, we are able to compute the exact marginalization over all alignments. Our models achieve state-of-the-art performance on morphological inflection. Furthermore, we find strong performance on two other character-level transduction tasks. Code is available at https://github.com/shijie-wu/neural-transducer.

We present a study of morphological irregularity. Following recent work, we define an information-theoretic measure of irregularity based on the predictability of forms in a language. Using a neural transduction model, we estimate this quantity for the forms in 28 languages. We first present several validatory and exploratory analyses of irregularity. We then show that our analyses provide evidence for a correlation between irregularity and frequency: higher frequency items are more likely to be irregular and irregular items are more likely be highly frequent. To our knowledge, this result is the first of its breadth and confirms longstanding proposals from the linguistics literature. The correlation is more robust when aggregated at the level of whole paradigms—providing support for models of linguistic structure in which inflected forms are unified by abstract underlying stems or lexemes.

The SIGMORPHON 2019 shared task on cross-lingual transfer and contextual analysis in morphology examined transfer learning of inflection between 100 language pairs, as well as contextual lemmatization and morphosyntactic description in 66 languages. The first task evolves past years’ inflection tasks by examining transfer of morphological inflection knowledge from a high-resource language to a low-resource language. This year also presents a new second challenge on lemmatization and morphological feature analysis in context. All submissions featured a neural component and built on either this year’s strong baselines or highly ranked systems from previous years’ shared tasks. Every participating team improved in accuracy over the baselines for the inflection task (though not Levenshtein distance), and every team in the contextual analysis task improved on both state-of-the-art neural and non-neural baselines.

Character-level string-to-string transduction is an important component of various NLP tasks. The goal is to map an input string to an output string, where the strings may be of different lengths and have characters taken from different alphabets. Recent approaches have used sequence-to-sequence models with an attention mechanism to learn which parts of the input string the model should focus on during the generation of the output string. Both soft attention and hard monotonic attention have been used, but hard non-monotonic attention has only been used in other sequence modeling tasks and has required a stochastic approximation to compute the gradient. In this work, we introduce an exact, polynomial-time algorithm for marginalizing over the exponential number of non-monotonic alignments between two strings, showing that hard attention models can be viewed as neural reparameterizations of the classical IBM Model 1. We compare soft and hard non-monotonic attention experimentally and find that the exact algorithm significantly improves performance over the stochastic approximation and outperforms soft attention.