The current focus on resource-rich languages poses a challenge to linguistic diversity, affecting minority languages with limited digital presence and relatively old published and unpublished resources. In addressing this issue, this study targets the digitalization of old scanned textbooks written in four Peruvian indigenous languages (Asháninka, Shipibo-Konibo, Yanesha, and Yine) using Optical Character Recognition (OCR) technology. This is complemented with text correction methods to minimize extraction errors. Contributions include the creation of an annotated dataset with 454 scanned page images, for a rigorous evaluation, and the development of a module to correct OCR-generated transcription alignments.

We introduce a Spanish-Awajun parallel dataset of 22k high-quality sentence pairs with the help of the journalistic organization Company C. This dataset consists of parallel data obtained from various web sources such as poems, stories, laws, protocols, guidelines, handbooks, the Bible, and news published by Company C. The study also includes an analysis of the dataset’s performance for Spanish-Awajun translation using a Transformer architecture with transfer learning from a parent model, utilizing Spanish-English and Spanish-Finnish as high-resource language-pairs. As far as we know, this is the first Spanish-Awajun machine translation study, and we hope that this work will serve as a starting point for future research on this neglected Peruvian language.

This paper presents the results of the first shared task about the creation of educational materials for three indigenous languages of the Americas.The task proposes to automatically generate variations of sentences according to linguistic features that could be used for grammar exercises.The languages involved in this task are Bribri, Maya, and Guarani.Seven teams took part in the challenge, submitting a total of 22 systems, obtaining very promising results.

This paper presents the findings of the third iteration of the AmericasNLP Shared Task on Machine Translation. This year’s competition features eleven Indigenous languages found across North, Central, and South America. A total of six teams participate with a total of 157 submissions across all languages and models. Two baselines – the Sheffield and Helsinki systems from 2023 – are provided and represent hard-to-beat starting points for the competition. In addition to the baselines, teams are given access to a new repository of training data which consists of data collected by teams in prior shared tasks. Using ChrF++ as the main competition metric, we see improvements over the baseline for 4 languages: Chatino, Guarani, Quechua, and Rarámuri, with performance increases over the best baseline of 4.2 ChrF++. In this work, we present a summary of the submitted systems, results, and a human evaluation of system outputs for Bribri, which consists of both (1) a rating of meaning and fluency and (2) a qualitative error analysis of outputs from the best submitted system.

Multilingual pretraining approaches in Neural Machine Translation (NMT) have shown that training models to denoise synthetic code-switched data can yield impressive performance gains — owing to better multilingual semantic representations and transfer learning. However, they generated the synthetic code-switched data using non-contextual, one-to-one word translations obtained from lexicons - which can lead to significant noise in a variety of cases, including the poor handling of polysemes and multi-word expressions, violation of linguistic agreement and inability to scale to agglutinative languages. To overcome these limitations, we propose an approach called Contextual Code-Switching (CCS), where contextual, many-to-many word translations are generated using a ‘base’ NMT model. We conduct experiments on 3 different language families - Romance, Uralic, and Indo-Aryan - and show significant improvements (by up to 5.5 spBLEU points) over the previous lexicon-based SOTA approaches. We also observe that small CCS models can perform comparably or better than massive models like mBART50 and mRASP2, depending on the size of data provided. We empirically analyse several key factors responsible for these - including context, many-to-many substitutions, code-switching language count etc. - and prove that they all contribute to enhanced pretraining of multilingual NMT models.

Large multilingual models have inspired a new class of word alignment methods, which work well for the model’s pretraining languages. However, the languages most in need of automatic alignment are low-resource and, thus, not typically included in the pretraining data. In this work, we ask: How do modern aligners perform on unseen languages, and are they better than traditional methods? We contribute gold-standard alignments for Bribri–Spanish, Guarani–Spanish, Quechua–Spanish, and Shipibo-Konibo–Spanish. With these, we evaluate state-of-the-art aligners with and without model adaptation to the target language. Finally, we also evaluate the resulting alignments extrinsically through two downstream tasks: named entity recognition and part-of-speech tagging. We find that although transformer-based methods generally outperform traditional models, the two classes of approach remain competitive with each other.

In this work, we present the results of the AmericasNLP 2023 Shared Task on Machine Translation into Indigenous Languages of the Americas. This edition of the shared task featured eleven language pairs, one of which – Chatino-Spanish – uses a newly collected evaluation dataset, consisting of professionally translated text from the legal domain. Seven teams participated in the shared task, with a total of 181 submissions. Additionally, we conduct a human evaluation of the best system outputs, and compare them to the best submissions from the prior shared task. We find that this analysis agrees with the quantitative measures used to rank submissions, which shows further improvements of 9.64 ChrF on average across all languages, when compared to the prior winning system.

Theoretical work in morphological typology offers the possibility of measuring morphological diversity on a continuous scale. However, literature in Natural Language Processing (NLP) typically labels a whole language with a strict type of morphology, e.g. fusional or agglutinative. In this work, we propose to reduce the rigidity of such claims, by quantifying morphological typology at the word and segment level. We consider Payne (2017)’s approach to classify morphology using two indices: synthesis (e.g. analytic to polysynthetic) and fusion (agglutinative to fusional). For computing synthesis, we test unsupervised and supervised morphological segmentation methods for English, German and Turkish, whereas for fusion, we propose a semi-automatic method using Spanish as a case study. Then, we analyse the relationship between machine translation quality and the degree of synthesis and fusion at word (nouns and verbs for English-Turkish, and verbs in English-Spanish) and segment level (previous language pairs plus English-German in both directions). We complement the word-level analysis with human evaluation, and overall, we observe a consistent impact of both indexes on machine translation quality.

Pretrained multilingual models are able to perform cross-lingual transfer in a zero-shot setting, even for languages unseen during pretraining. However, prior work evaluating performance on unseen languages has largely been limited to low-level, syntactic tasks, and it remains unclear if zero-shot learning of high-level, semantic tasks is possible for unseen languages. To explore this question, we present AmericasNLI, an extension of XNLI (Conneau et al., 2018) to 10 Indigenous languages of the Americas. We conduct experiments with XLM-R, testing multiple zero-shot and translation-based approaches. Additionally, we explore model adaptation via continued pretraining and provide an analysis of the dataset by considering hypothesis-only models. We find that XLM-R’s zero-shot performance is poor for all 10 languages, with an average performance of 38.48%. Continued pretraining offers improvements, with an average accuracy of 43.85%. Surprisingly, training on poorly translated data by far outperforms all other methods with an accuracy of 49.12%.

Morphologically-rich polysynthetic languages present a challenge for NLP systems due to data sparsity, and a common strategy to handle this issue is to apply subword segmentation. We investigate a wide variety of supervised and unsupervised morphological segmentation methods for four polysynthetic languages: Nahuatl, Raramuri, Shipibo-Konibo, and Wixarika. Then, we compare the morphologically inspired segmentation methods against Byte-Pair Encodings (BPEs) as inputs for machine translation (MT) when translating to and from Spanish. We show that for all language pairs except for Nahuatl, an unsupervised morphological segmentation algorithm outperforms BPEs consistently and that, although supervised methods achieve better segmentation scores, they under-perform in MT challenges. Finally, we contribute two new morphological segmentation datasets for Raramuri and Shipibo-Konibo, and a parallel corpus for Raramuri–Spanish.

The Universal Morphology (UniMorph) project is a collaborative effort providing broad-coverage instantiated normalized morphological inflection tables for hundreds of diverse world languages. The project comprises two major thrusts: a language-independent feature schema for rich morphological annotation, and a type-level resource of annotated data in diverse languages realizing that schema. This paper presents the expansions and improvements on several fronts that were made in the last couple of years (since McCarthy et al. (2020)). Collaborative efforts by numerous linguists have added 66 new languages, including 24 endangered languages. We have implemented several improvements to the extraction pipeline to tackle some issues, e.g., missing gender and macrons information. We have amended the schema to use a hierarchical structure that is needed for morphological phenomena like multiple-argument agreement and case stacking, while adding some missing morphological features to make the schema more inclusive. In light of the last UniMorph release, we also augmented the database with morpheme segmentation for 16 languages. Lastly, this new release makes a push towards inclusion of derivational morphology in UniMorph by enriching the data and annotation schema with instances representing derivational processes from MorphyNet.

In this paper, we launch a new Universal Dependencies treebank for an endangered language from Amazonia: Kakataibo, a Panoan language spoken in Peru. We first discuss the collaborative methodology implemented, which proved effective to create a treebank in the context of a Computational Linguistic course for undergraduates. Then, we describe the general details of the treebank and the language-specific considerations implemented for the proposed annotation. We finally conduct some experiments on part-of-speech tagging and syntactic dependency parsing. We focus on monolingual and transfer learning settings, where we study the impact of a Shipibo-Konibo treebank, another Panoan language resource.

Spell-checkers are core applications in language learning and normalisation, which may enormously contribute to language revitalisation and language teaching in the context of indigenous communities. Spell-checking as a generation task, however, requires large amount of data, which is not feasible for endangered languages, such as the languages spoken in Peruvian Amazonia. We propose here augmentation methods for various misspelling types as a strategy to train neural spell-checking models and we create an evaluation resource for four indigenous languages of Peru: Shipibo-Konibo, Asháninka, Yánesha, Yine. We focus on special errors that are significant for learning these languages, such as phoneme-to-grapheme ambiguity, grammatical errors (gender, tense, number, among others), accentuation, punctuation and normalisation in contexts where two or more writing traditions co-exist. We found that an ensemble model, trained with augmented data from various types of error achieves overall better scores in most of the error types and languages. Finally, we released our spell-checkers as a web service to be used by indigenous communities and organisations to develop future language materials.

Language revitalisation should not be understood as a direct outcome of language documentation, which is mainly focused on the creation of language repositories. Natural language processing (NLP) offers the potential to complement and exploit these repositories through the development of language technologies that may contribute to improving the vitality status of endangered languages. In this paper, we discuss the current state of the interaction between language documentation and computational linguistics, present a diagnosis of how the outputs of recent documentation projects for endangered languages are underutilised for the NLP community, and discuss how the situation could change from both the documentary linguistics and NLP perspectives. All this is introduced as a bridging paradigm dubbed as Computational Language Documentation and Development (CLD²). CLD² calls for (1) the inclusion of NLP-friendly annotated data as a deliverable of future language documentation projects; and (2) the exploitation of language documentation databases by the NLP community to promote the computerization of endangered languages, as one way to contribute to their revitalization.

Language modelling and machine translation tasks mostly use subword or character inputs, but syllables are seldom used. Syllables provide shorter sequences than characters, require less-specialised extracting rules than morphemes, and their segmentation is not impacted by the corpus size. In this study, we first explore the potential of syllables for open-vocabulary language modelling in 21 languages. We use rule-based syllabification methods for six languages and address the rest with hyphenation, which works as a syllabification proxy. With a comparable perplexity, we show that syllables outperform characters and other subwords. Moreover, we study the importance of syllables on neural machine translation for a non-related and low-resource language-pair (Spanish–Shipibo-Konibo). In pairwise and multilingual systems, syllables outperform unsupervised subwords, and further morphological segmentation methods, when translating into a highly synthetic language with a transparent orthography (Shipibo-Konibo). Finally, we perform some human evaluation, and discuss limitations and opportunities.

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.

We represent the complexity of Yine (Arawak) morphology with a finite state transducer (FST) based morphological analyzer. Yine is a low-resource indigenous polysynthetic Peruvian language spoken by approximately 3,000 people and is classified as ‘definitely endangered’ by UNESCO. We review Yine morphology focusing on morphophonology, possessive constructions and verbal predicates. Then we develop FSTs to model these components proposing techniques to solve challenging problems such as complex patterns of incorporating open and closed category arguments. This is a work in progress and we still have more to do in the development and verification of our analyzer. Our analyzer will serve both as a tool to better document the Yine language and as a component of natural language processing (NLP) applications such as spell checking and correction.

Peru is a multilingual country with a long history of contact between the indigenous languages and Spanish. Taking advantage of this context for machine translation is possible with multilingual approaches for learning both unsupervised subword segmentation and neural machine translation models. The study proposes the first multilingual translation models for four languages spoken in Peru: Aymara, Ashaninka, Quechua and Shipibo-Konibo, providing both many-to-Spanish and Spanish-to-many models and outperforming pairwise baselines in most of them. The task exploited a large English-Spanish dataset for pre-training, monolingual texts with tagged back-translation, and parallel corpora aligned with English. Finally, by fine-tuning the best models, we also assessed the out-of-domain capabilities in two evaluation datasets for Quechua and a new one for Shipibo-Konibo.

This paper presents the results of the 2021 Shared Task on Open Machine Translation for Indigenous Languages of the Americas. The shared task featured two independent tracks, and participants submitted machine translation systems for up to 10 indigenous languages. Overall, 8 teams participated with a total of 214 submissions. We provided training sets consisting of data collected from various sources, as well as manually translated sentences for the development and test sets. An official baseline trained on this data was also provided. Team submissions featured a variety of architectures, including both statistical and neural models, and for the majority of languages, many teams were able to considerably improve over the baseline. The best performing systems achieved 12.97 ChrF higher than baseline, when averaged across languages.

We describe the University of Edinburgh’s submissions to the WMT20 news translation shared task for the low resource language pair English-Tamil and the mid-resource language pair English-Inuktitut. We use the neural machine translation transformer architecture for all submissions and explore a variety of techniques to improve translation quality to compensate for the lack of parallel training data. For the very low-resource English-Tamil, this involves exploring pretraining, using both language model objectives and translation using an unrelated high-resource language pair (German-English), and iterative backtranslation. For English-Inuktitut, we explore the use of multilingual systems, which, despite not being part of the primary submission, would have achieved the best results on the test set.

We introduce new monolingual corpora for four indigenous and endangered languages from Peru: Shipibo-konibo, Ashaninka, Yanesha and Yine. Given the total absence of these languages in the web, the extraction and processing of texts from PDF files is relevant in a truly low-resource language scenario. Our procedure for monolingual corpus creation considers language-specific and language-agnostic steps, and focuses on educational PDF files with multilingual sentences, noisy pages and low-structured content. Through an evaluation based on language modelling and character-level perplexity on a subset of manually extracted sentences, we determine that our method allows the creation of clean corpora for the four languages, a key resource for natural language processing tasks nowadays.

In hierarchical text classification, we perform a sequence of inference steps to predict the category of a document from top to bottom of a given class taxonomy. Most of the studies have focused on developing novels neural network architectures to deal with the hierarchical structure, but we prefer to look for efficient ways to strengthen a baseline model. We first define the task as a sequence-to-sequence problem. Afterwards, we propose an auxiliary synthetic task of bottom-up-classification. Then, from external dictionaries, we retrieve textual definitions for the classes of all the hierarchy’s layers, and map them into the word vector space. We use the class-definition embeddings as an additional input to condition the prediction of the next layer and in an adapted beam search. Whereas the modified search did not provide large gains, the combination of the auxiliary task and the additional input of class-definitions significantly enhance the classification accuracy. With our efficient approaches, we outperform previous studies, using a drastically reduced number of parameters, in two well-known English datasets.

Sparse language vectors from linguistic typology databases and learned embeddings from tasks like multilingual machine translation have been investigated in isolation, without analysing how they could benefit from each other’s language characterisation. We propose to fuse both views using singular vector canonical correlation analysis and study what kind of information is induced from each source. By inferring typological features and language phylogenies, we observe that our representations embed typology and strengthen correlations with language relationships. We then take advantage of our multi-view language vector space for multilingual machine translation, where we achieve competitive overall translation accuracy in tasks that require information about language similarities, such as language clustering and ranking candidates for multilingual transfer. With our method, we can easily project and assess new languages without expensive retraining of massive multilingual or ranking models, which are major disadvantages of related approaches.

We introduce a shift on the DS method over the domain of crime-related news from Peru, attempting to find the culprit, victim and location of a crime description from a RE perspective. Obtained results are highly promising and show that proposed modifications are effective in non-traditional domains.

Corpora curated by experts have sustained Natural Language Processing mainly in English, but the expensiveness of corpora creation is a barrier for the development in further languages. Thus, we propose a corpus generation strategy that only requires a machine translation system between English and the target language in both directions, where we filter the best translations by computing automatic translation metrics and the task performance score. By studying Reading Comprehension in Spanish and Word Sense Disambiguation in Portuguese, we identified that a more quality-oriented metric has high potential in the corpora selection without degrading the task performance. We conclude that it is possible to systematise the building of quality corpora using machine translation and automatic metrics, besides some prior effort to clean and process the data.

We present an initial version of the Universal Dependencies (UD) treebank for Shipibo-Konibo, the first South American, Amazonian, Panoan and Peruvian language with a resource built under UD. We describe the linguistic aspects of how the tagset was defined and the treebank was annotated; in addition we present our specific treatment of linguistic units called clitics. Although the treebank is still under development, it allowed us to perform a typological comparison against Spanish, the predominant language in Peru, and dependency syntax parsing experiments in both monolingual and cross-lingual approaches.

In this paper, we present the first attempts to develop a machine translation (MT) system between Spanish and Shipibo-konibo (es-shp). There are very few digital texts written in Shipibo-konibo and even less bilingual texts that can be aligned, hence we had to create a parallel corpus using both bilingual and monolingual texts. We will describe how this corpus was made, as well as the process we followed to improve the quality of the sentences used to build a statistical MT model or SMT. The results obtained surpassed the baseline proposed (dictionary based) and made a promising result for further development considering the size of corpus used. Finally, it is expected that this MT system can be reinforced with the use of additional linguistic rules and automatic language processing functions that are being implemented.

There are several native languages in Peru which are mostly agglutinative. These languages are transmitted from generation to generation mainly in oral form, causing different forms of writing across different communities. For this reason, there are recent efforts to standardize the spelling in the written texts, and it would be beneficial to support these tasks with an automatic tool such as an spell-checker. In this way, this spelling corrector is being developed based on two steps: an automatic rule-based syllabification method and a character-level graph to detect the degree of error in a misspelled word. The experiments were realized on Shipibo-konibo, a highly agglutinative and amazonian language, and the results obtained have been promising in a dataset built for the purpose.