This paper presents our submission to the AmericasNLP 2024 Shared Task on the Creation of Educational Materials for Indigenous Languages. We frame this task as one of morphological inflection generation, treating each sentence as a single word. We investigate and compare two distinct approaches: fine-tuning neural encoder-decoder models such as NLLB- 200, and in-context learning with proprietary large language models (LLMs). Our findings demonstrate that for this task, no one approach is perfect. Anthropic’s Claude 3 Opus, when supplied with grammatical description entries, achieves the highest performance on Bribri among the evaluated models. This outcome corroborates and extends previous research exploring the efficacy of in-context learning in low- resource settings. For Maya, fine-tuning NLLB- 200-3.3B using StemCorrupt augmented data yielded the best performance.

This paper describes a curriculum for teaching linguists how to apply machine-in-the-loop (MitL) approach to documentary and descriptive tasks. It also shares observations about the learning participants, who are primarily non-computational linguists, and how they interact with the MitL approach. We found that they prefer cleaning over increasing the training data and then proceed to reanalyze their analytical decisions, before finally undertaking small actions that emphasize analytical strategies. Overall, participants display an understanding of the curriculum which covers fundamental concepts of machine learning and statistical modeling.

We contribute a seed dataset for the Bangla/Bengali language as part of the WMT24 Open Language Data Initiative shared task. We validate the quality of the dataset against a mined and automatically aligned dataset (NLLBv1) and two other existing datasets of crowdsourced manual translations. The validation is performed by investigating the performance of state-of-the-art translation models fine-tuned on the different datasets after controlling for training set size. Machine translation models fine-tuned on our dataset outperform models tuned on the other datasets in both translation directions (English-Bangla and Bangla-English). These results confirm the quality of our dataset. We hope our dataset will support machine translation for the Bangla/Bengali community and related low-resource languages.

African American English (AAE) has received recent attention in the field of natural language processing (NLP). Efforts to address bias against AAE in NLP systems tend to focus on lexical differences. When the unique structures of AAE are considered, the solution is often to remove or neutralize the differences. This work leverages knowledge about the unique linguistic structures to improve automatic disambiguation of habitual and non-habitual meanings of “be” in naturally produced AAE transcribed speech. Both meanings are employed in AAE but examples of Habitual be are rare in already limited AAE data. Generally, representing additional syntactic information improves semantic disambiguation of habituality. Using an ensemble of classical machine learning models with a representation of the unique POS and dependency patterns of Habitual be, we show that integrating syntactic information improves the identification of habitual uses of “be” by about 65 F1 points over a simple baseline model of n-grams, and as much as 74 points. The success of this approach demonstrates the potential impact when we embrace, rather than neutralize, the structural uniqueness of African American English.

This paper presents the findings of the SIGMORPHON 2023 Shared Task on Interlinear Glossing. This first iteration of the shared task explores glossing of a set of six typologically diverse languages: Arapaho, Gitksan, Lezgi, Natügu, Tsez and Uspanteko. The shared task encompasses two tracks: a resource-scarce closed track and an open track, where participants are allowed to utilize external data resources. Five teams participated in the shared task. The winning team Tü-CL achieved a 23.99%-point improvement over a baseline RoBERTa system in the closed track and a 17.42%-point improvement in the open track.

Recent research has highlighted that natural language processing (NLP) systems exhibit a bias againstAfrican American speakers. These errors are often caused by poor representation of linguistic features unique to African American English (AAE), which is due to the relatively low probability of occurrence for many such features. We present a workflow to overcome this issue in the case of habitual “be”. Habitual “be” is isomorphic, and therefore ambiguous, with other forms of uninflected “be” found in both AAE and General American English (GAE). This creates a clear challenge for bias in NLP technologies. To overcome the scarcity, we employ a combination of rule-based filters and data augmentation that generate a corpus balanced between habitual and non-habitual instances. This balanced corpus trains unbiased machine learning classifiers, as demonstrated on a corpus of AAE transcribed texts, achieving .65 F₁ score at classifying habitual “be”.

Part-of-Speech (POS) tags are routinely included as features in many NLP tasks. However, the importance and usefulness of POS tags needs to be examined as NLP expands to low-resource languages because linguists who provide many annotated resources do not place priority on early identification and tagging of POS. This paper describes an empirical study about the effect that POS tags have on two computational morphological tasks with the Transformer architecture. Each task is tested twice on identical data except for the presence/absence of POS tags, using published data in ten high- to low-resource languages or unpublished linguistic field data in five low-resource languages. We find that the presence or absence of POS tags does not have a significant bearing on performance. In joint segmentation and glossing, the largest average difference is an .09 improvement in F1-scores by removing POS tags. In reinflection, the greatest average difference is 1.2% in accuracy for published data and 5% for unpublished and noisy field data.

This paper presents a proposition bank for Russian (RuPB), a resource for semantic role labeling (SRL). The motivating goal for this resource is to automatically project semantic role labels from English to Russian. This paper describes frame creation strategies, coverage, and the process of sense disambiguation. It discusses language-specific issues that complicated the process of building the PropBank and how these challenges were exploited as language-internal guidance for consistency and coherence.

Despite recent advances in natural language processing and other language technology, the application of such technology to language documentation and conservation has been limited. In August 2019, a workshop was held at Carnegie Mellon University in Pittsburgh, PA, USA to attempt to bring together language community members, documentary linguists, and technologists to discuss how to bridge this gap and create prototypes of novel and practical language revitalization technologies. The workshop focused on developing technologies to aid language documentation and revitalization in four areas: 1) spoken language (speech transcription, phone to orthography decoding, text-to-speech and text-speech forced alignment), 2) dictionary extraction and management, 3) search tools for corpora, and 4) social media (language learning bots and social media analysis). This paper reports the results of this workshop, including issues discussed, and various conceived and implemented technologies for nine languages: Arapaho, Cayuga, Inuktitut, Irish Gaelic, Kidaw’ida, Kwak’wala, Ojibwe, San Juan Quiahije Chatino, and Seneca.

An intermediate step in the linguistic analysis of an under-documented language is to find and organize inflected forms that are attested in natural speech. From this data, linguists generate unseen inflected word forms in order to test hypotheses about the language’s inflectional patterns and to complete inflectional paradigm tables. To get the data linguists spend many hours manually creating interlinear glossed texts (IGTs). We introduce a new task that speeds this process and automatically generates new morphological resources for natural language processing systems: IGT-to-paradigms (IGT2P). IGT2P generates entire morphological paradigms from IGT input. We show that existing morphological reinflection models can solve the task with 21% to 64% accuracy, depending on the language. We further find that (i) having a language expert spend only a few hours cleaning the noisy IGT data improves performance by as much as 21 percentage points, and (ii) POS tags, which are generally considered a necessary part of NLP morphological reinflection input, have no effect on the accuracy of the models considered here.

In the field of metaphor detection, deep learning systems are the ubiquitous and achieve strong performance on many tasks. However, due to the complicated procedures for manually identifying metaphors, the datasets available are relatively small and fraught with complications. We show that using syntactic features and lexical resources can automatically provide additional high-quality training data for metaphoric language, and this data can cover gaps and inconsistencies in metaphor annotation, improving state-of-the-art word-level metaphor identification. This novel application of automatically improving training data improves classification across numerous tasks, and reconfirms the necessity of high-quality data for deep learning frameworks.

Semantic relations are often signaled with prepositional or possessive marking—but extreme polysemy bedevils their analysis and automatic interpretation. We introduce a new annotation scheme, corpus, and task for the disambiguation of prepositions and possessives in English. Unlike previous approaches, our annotations are comprehensive with respect to types and tokens of these markers; use broadly applicable supersense classes rather than fine-grained dictionary definitions; unite prepositions and possessives under the same class inventory; and distinguish between a marker’s lexical contribution and the role it marks in the context of a predicate or scene. Strong interannotator agreement rates, as well as encouraging disambiguation results with established supervised methods, speak to the viability of the scheme and task.

We experiment with training an encoder-decoder neural model for mimicking the behavior of an existing hand-written finite-state morphological grammar for Arapaho verbs, a polysynthetic language with a highly complex verbal inflection system. After adjusting for ambiguous parses, we find that the system is able to generalize to unseen forms with accuracies of 98.68% (unambiguous verbs) and 92.90% (all verbs).

Morphological analysis of morphologically rich and low-resource languages is important to both descriptive linguistics and natural language processing. Field documentary efforts usually procure analyzed data in cooperation with native speakers who are capable of providing some level of linguistic information. Manually annotating such data is very expensive and the traditional process is arguably too slow in the face of language endangerment and loss. We report on a case study of learning to automatically gloss a Nakh-Daghestanian language, Lezgi, from a very small amount of seed data. We compare a conditional random field based sequence labeler and a neural encoder-decoder model and show that a nearly 0.9 F1-score on labeled accuracy of morphemes can be achieved with 3,000 words of transcribed oral text. Errors are mostly limited to morphemes with high allomorphy. These results are potentially useful for developing rapid annotation and fieldwork tools to support documentation of morphologically rich, endangered languages.