A POS Tier Is the Key to Automated Annotation for Low-Resource Language Documentation: Neural Interlinear Glossing for Irabu, a Southern Ryukyuan Language

Parsing Middle High German: exploring cross-lingual NLP for treebank construction in low-resource historical languages

Building syntactically annotated corpora, such as treebanks, for historical languages is a challenging yet vital task in digital humanities, as it underpins linguistic analysis and facilitates a range of interdisciplinary research. However, the scarcity of annotated data and the need for extensive expertise in historical linguistics make this process particularly demanding. In this study, we explore the potential of cross-lingual natural language processing (NLP) techniques as a semiautomatic solution for treebank construction in low-resource historical languages. We use Middle High German (MHG) as a case study. Leveraging the linguistic continuity and structural similarities between MHG and Modern German (MG), we effectively utilize the extensive MG treebank resources to develop a constituency parsing system tailored for MHG. Specifically, to design a semiautomatic system that integrates automatic annotation with manual validation, we explore two cross-lingual transfer techniques: zero-shot transfer and delexicalization; the latter removes lexical information to focus on syntactic structure. In our experiments, we first train parsers on MG treebanks, and then transfer them to MHG using the two cross-lingual transfer techniques. The delexicalization method achieves a parsing performance of 67.3 per cent in terms of F1-score. This performance significantly surpasses the zero-shot cross-lingual method by a margin of 28.6 percentage points. These investigations validate the effectiveness and feasibility of cross-lingual transfer techniques for historical language treebank construction. This study highlights the potential of NLP tools to streamline the semiautomatic annotation process, reducing the reliance on extensive linguistic expertise and manual effort, and paving the way for broader applications in digital humanities research.

Minimal-Supervision Part-of-Speech Tagging for Assamese Language: An Evidence-Gated Cascade with Classifier Backfill

Part-of-speech (POS) tagging for low-resource languages is limited more by the expense of generating labeled training data than by algorithmic constraints. This study evaluates the extent to which a length-stratified, minimally-supervised pipeline can recover tagging accuracy for the Assamese language using a labeled seed of only 3 to 5 sentences per distinct sentence length (142 to 228 sentences in total), compared to a weak-supervision classifier trained on a conventional 80% corpus split (2,840 sentences). An evidence-gated cascade is constructed, comprising Brown-style word clustering, a seed-ambiguity audit with context-based disambiguation, confidence-gated cluster labeling, and empirically validated rule fallbacks. This approach achieves 94.9% to 95.0% accuracy but only 19.8% to 20.0% token coverage from its minimal seed. To address the coverage gap, every remaining token is backfilled using a classifier trained on the pipeline’s accumulated evidence, resulting in 100% coverage at 84.7% to 86.0% accuracy (depending on seed size), as evaluated against a fixed, shared test set for direct comparison with the larger-budget baseline (90.8% accuracy, same coverage, same test set). Three negative results are reported alongside the positive findings: Brown-style clustering provides negligible benefit at this scale, Viterbi sequence decoding does not transfer to the backfill classifier from the larger-budget setting, and naive self-training fails due to confirmation bias. Additionally, through five replicated random seed draws per condition, it is demonstrated that the residual seed-size effect (3 versus 5 sentences per length, closing approximately 21% of the accuracy gap to the larger-budget baseline) is statistically significant rather than the result of a single favorable draw.

Ensemble-Based Approach for Amazigh POS Tagging: Leveraging Multiple Models for Enhanced Performance in Low-Resource Language Processing

Part-of-Speech (POS) tagging is a foundational task in Natural Language Processing (NLP), yet it remains challenging for low-resource and morphologically rich languages such as Amazigh. This paper proposes a hybrid ensemble framework for Amazigh POS tagging that integrates three complementary models: a Bidirectional Long Short-Term Memory network (BiLSTM), a Conditional Random Field model (CRF), and a rule-based morphological analyzer (RBMA). Rather than treating prior results obtained on different corpora and tag inventories as directly comparable, the study evaluates all proposed components under a common 54-tag experimental setting based on the publicly available Amazigh Linguistic Dataset. Three ensemble strategies are examined: majority voting, validation-weighted voting, and logistic-regression stacking. An additional late-fusion ablation applies hard and soft RBMA constraints to CRF and Stacking outputs; hard masking degrades performance substantially, whereas soft masking is more robust but remains below unconstrained decoding. The best micro-level performance is obtained by the stacking ensemble, which reaches 98.51% Micro-F1/accuracy, whereas the boosting-like weighted ensemble obtains the strongest Macro-F1 among the ensemble variants, reaching 74.24%. These results show that hybrid ensemble methods can improve token-level accuracy in low-resource POS tagging, while also revealing a trade-off between frequent-tag accuracy and rare-tag robustness. The findings highlight the usefulness of combining neural, probabilistic, and rule-based information for Amazigh POS tagging, and point to class-balanced meta-learning and character/subword representations as important directions for improving rare and out-of-vocabulary categories.

Leveraging LLMs to Automatically Construct WordNets as Bilingual Resources

From transcription to semantic corpus analysis: unsupervised learning of sentence representations for ancient languages

Automatic Text Recognition (ATR) now supplies digital humanities with large volumes of unstructured, heterogeneous, and often noisy text in ancient languages. Downstream semantic analysestext reuse identification, alignment, and semantic search-rely on sentence embeddings, yet existing methods transfer poorly to ancient languages: generic multilingual encoders underperform, specialized language models yield anisotropic representation spaces, and labeled similarity data is unavailable. We study two fully unsupervised strategies - TSDAE and contrastive sentence embedding (CSE) - that adapt a specialized token-level language model into a corpus-specific sentence encoder using only raw sentences. On the philologically central case of biblical reuse in patristic literature (2,935 expert-verified parallels in Latin and Ancient Greek, from Augustine, Jerome, and Athanasius), we decompose reuse identification into two separately evaluated tasks-binary detection and correspondence retrieval-and benchmark the adapted encoders against multilingual, specialized, distilled, and supervised fine-tuned baselines, as well as on artificially noised data simulating HTR artifacts and scribal abbreviations. The adapted encoders outperform all baselines on both tasks, with complementary profiles: TSDAE leads detection given a large in-domain corpus, while CSE leads retrieval, reaches its optimum with as few as 4-8k raw in-domain sentences-a few tens of seconds of training on a laptop GPU-and transfers across works and authors, including to noisy post-ATR text when retrained directly on it. UMAP atlases relate the geometric effect of each strategy to the measured gains, and the full pipeline-segmentation, fine-tuning, cross-corpus semantic search-is made available to non-specialists through the online tool Paraphrasis.

Sahara Tokenizers at MWE-2026 PARSEME 2.0 Subtask 1: Combining Contextual Embeddings with Structural Decoding for Multi-Word Expression Detection

Alation studies reveal a strong synergy between POS features and CRF decoding, with the combined approach yielding the best single-model performance, and ensembling models trained with different objectives improves both overall F1 score and discontinuous MWE scores, demonstrating the importance of training diversity for capturing non-adjacent syntactic patterns.