This work investigates the in-context learning abilities of pretrained large language models (LLMs) when instructed to translate text from a low-resource language into a high-resource language as part of an automated machine translation pipeline. We conduct a set of experiments translating Southern Quechua to Spanish and examine the informativity of various types of information retrieved from a constrained database of digitized pedagogical materials (dictionaries and grammar lessons) and parallel corpora. Using both automatic and human evaluation of model output, we conduct ablation studies that manipulate (1) context type (morpheme translations, grammar descriptions, and corpus examples), (2) retrieval methods (automated vs. manual), and (3) model type. Our results suggest that even relatively small LLMs are capable of utilizing prompt context for zero-shot low-resource translation when provided a minimally sufficient amount of relevant linguistic information. However, the variable effects of prompt type, retrieval method, model type, and language community-specific factors highlight the limitations of using even the best LLMs as translation systems for the majority of the world’s 7,000+ languages and their speakers.

This paper explores how Generative Adversarial Networks (GANs) learn representations of phonological phenomena. We analyze how GANs encode contrastive and non-contrastive nasality in French and English vowels by applying the ciwGAN architecture (Begus, 2021). Begus claims that ciwGAN encodes linguistically meaningful representations with categorical variables in its latent space and manipulating the latent variables shows an almost one to one corresponding control of the phonological features in ciwGAN’s generated outputs. However, our results show an interactive effect of latent variables on the features in the generated outputs, which suggests the learned representations in neural networks are different from the phonological representations proposed by linguists. On the other hand, ciwGAN is able to distinguish contrastive and noncontrastive features in English and French by encoding them differently. Comparing the performance of GANs learning from different languages results in a better understanding of what language specific features contribute to developing language specific phonological representations. We also discuss the role of training data frequencies in phonological feature learning.

We investigate how well words in the polysynthetic language Inuktitut can be translated by combining dictionary definitions, without use of a neural machine translation model trained on parallel text. Such a translation system would allow natural language technology to benefit from resources designed for community use in a language revitalization or education program, rather than requiring a separate parallel corpus. We show that the text-to-text generation capabilities of GPT-3 allow it to perform this task with BLEU scores of up to 18.5. We investigate prompting GPT-3 to provide multiple translations, which can help slightly, and providing it with grammar information, which is mostly ineffective. Finally, we test GPT-3’s ability to derive morpheme definitions from whole-word translations, but find this process is prone to errors including hallucinations.

OSU’s inflection system is a transformer whose input is augmented with an analogical exemplar showing how to inflect a different word into the target cell. In addition, alignment-based heuristic features indicate how well the exemplar is likely to match the output. OSU’s scores substantially improve over the baseline transformer for instances where an exemplar is available, though not quite matching the challenge winner. In Part 2, the system shows a tendency to over-apply the majority pattern in English, but not Arabic.

There are many challenges in morphological fieldwork annotation, it heavily relies on segmentation and feature labeling (which have both practical and theoretical drawbacks), it’s time-intensive, and the annotator needs to be linguistically trained and may still annotate things inconsistently. We propose a workflow that relies on unsupervised and active learning grounded in Word-and-Paradigm morphology (WP). Machine learning has the potential to greatly accelerate the annotation process and allow a human annotator to focus on problematic cases, while the WP approach makes for an annotation system that is word-based and relational, removing the need to make decisions about feature labeling and segmentation early in the process and allowing speakers of the language of interest to participate more actively, since linguistic training is not necessary. We present a proof-of-concept for the first step of the workflow, in a realistic fieldwork setting, annotators can process hundreds of forms per hour.

We investigate how abstract processes like suffixation can be learned from morphological inflection task data using an analogical memory-based framework. In this framework, the inflection target form is specified by providing an example inflection of another word in the language. We show that this model is capable of near-baseline performance on the SigMorphon 2020 inflection challenge. Such a model can make predictions for unseen languages, allowing us to perform one-shot inflection on natural languages and investigate morphological transfer with synthetic probes. Accuracy for one-shot transfer can be unexpectedly high for some target languages (88% in Shona) and language families (53% across Romance). Probe experiments show that the model learns partially generalizable representations of prefixation, suffixation and reduplication, aiding its ability to transfer. We argue that the degree of generality of these process representations also helps to explain transfer results from previous research.

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.

Classical accounts of child language learning invoke memory limits as a pressure to discover sparse, language-like representations of speech, while more recent proposals stress the importance of prediction for language learning. In this study, we propose a broad-coverage unsupervised neural network model to test memory and prediction as sources of signal by which children might acquire language directly from the perceptual stream. Our model embodies several likely properties of real-time human cognition: it is strictly incremental, it encodes speech into hierarchically organized labeled segments, it allows interactive top-down and bottom-up information flow, it attempts to model its own sequence of latent representations, and its objective function only recruits local signals that are plausibly supported by human working memory capacity. We show that much phonemic structure is learnable from unlabeled speech on the basis of these local signals. We further show that remembering the past and predicting the future both contribute to the linguistic content of acquired representations, and that these contributions are at least partially complementary.

In this paper, we deploy binary stochastic neural autoencoder networks as models of infant language learning in two typologically unrelated languages (Xitsonga and English). We show that the drive to model auditory percepts leads to latent clusters that partially align with theory-driven phonemic categories. We further evaluate the degree to which theory-driven phonological features are encoded in the latent bit patterns, finding that some (e.g. [+-approximant]), are well represented by the network in both languages, while others (e.g. [+-spread glottis]) are less so. Together, these findings suggest that many reliable cues to phonemic structure are immediately available to infants from bottom-up perceptual characteristics alone, but that these cues must eventually be supplemented by top-down lexical and phonotactic information to achieve adult-like phone discrimination. Our results also suggest differences in degree of perceptual availability between features, yielding testable predictions as to which features might depend more or less heavily on top-down cues during child language acquisition.

Scholars in inter-disciplinary fields like the Digital Humanities are increasingly interested in semantic annotation of specialized corpora. Yet, under-resourced languages, imperfect or noisily structured data, and user-specific classification tasks make it difficult to meet their needs using off-the-shelf models. Manual annotation of large corpora from scratch, meanwhile, can be prohibitively expensive. Thus, we propose an active learning solution for named entity recognition, attempting to maximize a custom model’s improvement per additional unit of manual annotation. Our system robustly handles any domain or user-defined label set and requires no external resources, enabling quality named entity recognition for Humanities corpora where such resources are not available. Evaluating on typologically disparate languages and datasets, we reduce required annotation by 20-60% and greatly outperform a competitive active learning baseline.

We investigate the lexical network properties of the large phoneme inventory Southern African language Mangetti Dune !Xung as it compares to English and other commonly-studied languages. Lexical networks are graphs in which nodes (words) are linked to their minimal pairs; global properties of these networks are believed to mediate lexical access in the minds of speakers. We show that the network properties of !Xung are within the range found in previously-studied languages. By simulating data (”pseudolexicons”) with varying levels of phonotactic structure, we find that the lexical network properties of !Xung diverge from previously-studied languages when fewer phonotactic constraints are retained. We conclude that lexical network properties are representative of an underlying cognitive structure which is necessary for efficient word retrieval and that the phonotactics of !Xung may be shaped by a selective pressure which preserves network properties within this cognitively useful range.

We present the first unsupervised LSTM speech segmenter as a cognitive model of the acquisition of words from unsegmented input. Cognitive biases toward phonological and syntactic predictability in speech are rooted in the limitations of human memory (Baddeley et al., 1998); compressed representations are easier to acquire and retain in memory. To model the biases introduced by these memory limitations, our system uses an LSTM-based encoder-decoder with a small number of hidden units, then searches for a segmentation that minimizes autoencoding loss. Linguistically meaningful segments (e.g. words) should share regular patterns of features that facilitate decoder performance in comparison to random segmentations, and we show that our learner discovers these patterns when trained on either phoneme sequences or raw acoustics. To our knowledge, ours is the first fully unsupervised system to be able to segment both symbolic and acoustic representations of speech.

This paper describes our “breaker” submission to the 2017 EMNLP “Build It Break It” shared task on sentiment analysis. In order to cause the “builder” systems to make incorrect predictions, we edited items in the blind test data according to linguistically interpretable strategies that allow us to assess the ease with which the builder systems learn various components of linguistic structure. On the whole, our submitted pairs break all systems at a high rate (72.6%), indicating that sentiment analysis as an NLP task may still have a lot of ground to cover. Of the breaker strategies that we consider, we find our semantic and pragmatic manipulations to pose the most substantial difficulties for the builder systems.

Although spanning thousands of years and genres as diverse as liturgy, historiography, lyric and other forms of prose and poetry, the body of Latin texts is still relatively sparse compared to English. Data sparsity in Latin presents a number of challenges for traditional Named Entity Recognition techniques. Solving such challenges and enabling reliable Named Entity Recognition in Latin texts can facilitate many down-stream applications, from machine translation to digital historiography, enabling Classicists, historians, and archaeologists for instance, to track the relationships of historical persons, places, and groups on a large scale. This paper presents the first annotated corpus for evaluating Named Entity Recognition in Latin, as well as a fully supervised model that achieves over 90% F-score on a held-out test set, significantly outperforming a competitive baseline. We also present a novel active learning strategy that predicts how many and which sentences need to be annotated for named entities in order to attain a specified degree of accuracy when recognizing named entities automatically in a given text. This maximizes the productivity of annotators while simultaneously controlling quality.

Since the 18th century, the novel has been one of the defining forms of English writing, a mainstay of popular entertainment and academic criticism. Despite its importance, however, there are few computational studies of the large-scale structure of novels—and many popular representations for discourse modeling do not work very well for novelistic texts. This paper describes a high-level representation of plot structure which tracks the frequency of mentions of different characters, topics and emotional words over time. The representation can distinguish with high accuracy between real novels and artificially permuted surrogates; characters are important for eliminating random permutations, while topics are effective at distinguishing beginnings from ends.