NERBench-Chhattisgarh: A Multi-Family NER Dataset for Low-Resource Indic Languages

NerAxom: a BIO-tagged NER dataset and hybrid neural-rule framework for Assamese

Named entity recognition (NER) in low-resource, morphologically rich languages such as Assamese (ISO 639-3: asm) remains a significant challenge due to the scarcity of annotated corpora and the limited applicability of models designed for resource-rich languages. Existing Assamese NER resources suffer from critical limitations: WikiAnn provides broad language coverage but insufficient data volume for neural model training; AsNER, while a gold-standard corpus, supports only five entity categories and lacks a formal tagging scheme, restricting its utility for downstream tasks such as relation extraction and information retrieval. Furthermore, prior Assamese NER systems have relied predominantly on traditional tagging approaches and classical machine learning methods, with limited exploration of modern pre-trained language models and linguistically motivated post-processing strategies. To address these gaps, we present NerAxom, a BIO-tagged NER dataset for Assamese comprising 4,173 sentences and 106,046 tokens annotated across seven entity categories: Person (PER), Location (LOC), Organization (ORG), Date (DATE), Work_of_Art (WOA), Occupation (OCC), and Number (NUM). The dataset was independently annotated by two trained native speakers, achieving an inter-annotator agreement of κ=0.82 (Cohen’s Kappa), with disagreements resolved through expert linguist adjudication. We evaluate NerAxom using two modeling paradigms: (i) a BiLSTM–CRF model with an attention mechanism, tested with FastText, BERT, and MuRIL embeddings; and (ii) direct fine-tuning of the MuRIL transformer as a token classifier. Among embedding-based models, MuRIL yields the highest F1-score of 68%, outperforming FastText (62%) and BERT (64%). Fine-tuning MuRIL directly as a token classifier achieves an F1-score of 70%, establishing a competitive transformer baseline. To address entity misclassifications arising from Assamese morphological complexity, we further introduce a set of language-specific post-processing rules based on morphological suffixes and keyword cues. The hybrid system combining MuRIL embeddings in the BiLSTM–CRF+Attention architecture with these linguistic rules achieves an F1-score of 71% and an overall accuracy of 82% on Assamese Wikipedia biographical text, competitive with the fine-tuned MuRIL transformer (70% F1) and demonstrating that linguistically informed post-processing provides complementary gains over embedding-based neural baselines. The rule component yields the largest category-wise gains for LOC (+20 F1), WOA (+11 F1), and ORG (+9 F1). The NerAxom dataset is publicly available to support further NER research in Assamese and related low-resource Indic languages.

MameLoshnLM: Yiddish Language Model and Evaluation Benchmark

MameLoshnLM, the first open-source 8B-parameter language model built specifically for Yiddish, is presented, providing both a foundation for Yiddish NLP and a practical template for language model development in historically rich but digitally underrepresented languages.

PAP_NER: A large-scale vietnamese administrative named entity recognition corpus and hybrid deep learning architecture

Named Entity Recognition (NER) is fundamental for automating administrative document processing in digital government systems. However, Vietnamese NLP research faces a critical infrastructure gap: existing datasets focus on generic information extraction (news, medical) rather than domain-specific administrative text. We present PAP_NER, the first large-scale, gold-standard Vietnamese administrative NER corpus comprising 162,801 sentences with 205,807 entity annotations across five entity types critical for e-Government workflows: Agency (CQ), Legal Document (VBPL), Object (ĐT), Datetime (NG), and Quantity (SL). The dataset was constructed through a rigorous human-in-the-loop annotation pipeline, achieving an inter-annotator agreement of κ = 0.85. We demonstrate PAP_NER’s value through comprehensive benchmarking of an established hybrid deep learning architecture, PhoBERT-CRF, which couples monolingual Transformer embeddings (PhoBERT) with Conditional Random Fields for structured prediction. PhoBERT-CRF achieves 97.95% Micro F1-score on the PAP_NER test set, significantly outperforming established baselines: BiLSTM+CRF (+2.01%), multilingual XLM-RoBERTa (+2.52%), and pure Transformer approaches (+0.44%). Ablation analysis reveals that the CRF layer provides statistically significant improvements for structurally complex entities (VBPL: + 0.96%, p < 0.05, McNemar’s test). We release PAP_NER publicly (DOI: 10.5281/zenodo.18044019) under Creative Commons BY 4.0 license to support reproducibility and enable further research in Vietnamese administrative NLP. This work establishes a foundational dataset and methodology for addressing the Vietnamese government NER gap, with implications for low-resource language NLP research.

Santham : A Curated Sanskrit–Tamil Dataset with Anvaya and Segmentation for Building and Evaluating Machine Translation

Enhancing Readability of Telugu Text Summarization Using Multi-scale Attention and Bio-inspired Optimization

The proposed MLOA-MA-ASeqNet architecture, a Multi-scale Attention and Adaptive Sequence-to-Sequence Network whose hierarchical encoder operates simultaneously at word, phrase and sentence level granularity, achieves the highest average score across fluency, adequacy, coherence and readability.

Kannada Named Entity Recognition Using Deep Learning Techniques

Named Entity Recognition (NER) is a natural language processing task concerned with identifying mentions of named entities and classifying them according to a predefined set of categories. Despite the success of NER in domains, where such data is abundant it remains a formidable challenge for low-resource languages such as Kannada. In this paper we discuss the possible ways to approach NER for the Kannada language. We explore various research directions including rule-based methods statistical machine learning neural networks and transformers based tagging methodologies. We highlight the various challenges in achieving NER for such a language and propose a transformer based contextual tagging framework for labelling sequences. We propose to use mBERT IndicBERT and XLM-RoBERTa language models pretrained on target and other related Indic language corpora and further fine-tune these models for the NER task. We discuss various aspects for experimentation including data collection labelling data preparation methods data-splits evaluation metrics comparison with other models hyper parameter tuning entity-wise analysis and error analysis.