Padàrtha is the first ontology‑grounded fine‑grained Named Entity Recognition benchmark for Classical Sanskrit, built on the Mahêbhárata epic. Its tag set, derived from the Nyáya‑Vai’séka ontological system, contains 18 fine‑grained categories under 10 nodes and maps to five standard coarse tags, ensuring compatibility with existing benchmarks. The dataset includes over 12.6K expert‑annotated entries and 108,335 entity mentions across 73,632 verses, plus a 5,000‑verse test set designed to challenge rare mentions, and the study compares generative NER models to traditional architectures, noting performance drops at finer granularity and difficulties with unseen entities.
By Sujoy Sarkar, Pretam Ray, Paramhans Shah, Manoj Balaji Jagadeeshan, Akash Gairola, Arjuna S R, Pawan Goyal
arXiv:2607. 23319v1 Announce Type: cross Abstract: Standard subword tokenization algorithms such as Byte-Pair Encoding (BPE) and SentencePiece are trained predominantly on modern language corpora and produce inefficient segmentations when applied to classical Indian languages.
By Poornima Kumaresan, Pavithra Muruganantham, Lakshmi Rajendran, Santhosh Sivasubramani
The paper compares generative and encoder-based neural models for multilingual Named Entity Recognition (NER) across the eleven languages of the Naamapadam benchmark. Five classic model families, four decoder-only large language models fine‑tuned with LoRA and 4‑bit NF4 quantisation, and nine generative models in zero‑to‑5‑shot inference were evaluated under strict CoNLL span‑level metrics. Encoder-based models (mBERT and XLM‑R) achieved substantially higher F1 scores—up to 0.675 on Hindi—than any generative architecture, with gaps of 7.5–40 percentage points; the best few‑shot result reached only 28% of the encoder baseline. The study identifies three language clusters (encoder‑dominant, partial‑coverage, and failure‑zone) and offers deployment guidelines based on transfer learning and low‑resource NLP principles.
By Jakkala Mahesh, Jatavath Shravan Kumar, Komalla Shivani, Sujoy Sarkar
arXiv:2607. 06544v1 Announce Type: new Abstract: As Artificial Intelligence (AI) makes inroads into different parts of the Indian subcontinent, there is significant interest in studying how AI impacts the linguistic and cultural foundations of this civilization.
By Aparna Madva, Sharath Srivatsa, Srinath Srinivasa, Tulika Saha
arXiv:2608. 01935v1 Announce Type: cross Abstract: Prior work in Ancient Greek NLP relies on corpora that do not disambiguate the phonemic vowel length of alpha, iota, and ypsilon, together known as the dichrona.
By Albin Th\"orn Cleland, Eric Cullhed
arXiv:2608.30092v1 Announce Type: cross
Abstract: We present Arkios, a 1.04B-parameter dense transformer pretrained from scratch on 150B tokens of bilingual English-Nepali text, using a custom single...
By Sajal Regmi, Siddhartha Pudasaini, Chetan Phakami Pun
arXiv:2607. 28274v1 Announce Type: cross Abstract: Modern Greek is a richly inflected language, yet the language models built for it are evaluated mainly on factual knowledge, and no benchmark is dedicated to their inflectional competence.
By Ioakeim Perros, Cleopatra Papadopoulou, Ayoub Kirouane, Christos Petrocheilos
The paper evaluates whether diachronic word embeddings can track semantic change in Sanskrit, an ancient low‑resource language with complex phonological and morphological features. A 2.7‑million‑token corpus covering four canonical periods is processed with a neural sandhi splitter and lemmatizer, and per‑period embeddings are trained. Validation against a curated set of 21 historical shifts shows that 19 shifts align with philological expectations, supporting the method’s applicability to Sanskrit.
By Tanay Agrawal
Large language models (LLMs) process text through subword tokenizers rather than directly reading characters or words. Because these tokenizers are trained predominantly on English-centric corpora, they introduce a systematic and often overlooked disadvantage for many non-English languages.
The paper shows that HuggingFace’s ByteLevel pre‑tokenizer, which treats a word as a sequence of Unicode letters, splits abugida scripts at every vowel sign, creating a training‑free lower bound on tokenizer fertility. Across 26 languages, all 17 abugidas exhibit increased token counts (up to 9×), while Latin, Cyrillic, Hangul, and Han remain unchanged. The authors demonstrate that correcting the character class reduces Nepali token counts, improves model performance, and that this issue is widespread in popular HuggingFace models.
By Sajal Regmi, Siddhartha Pudasaini, Chetan Phakami Pun
arXiv:2606. 09767v1 Announce Type: cross Abstract: Neural machine translation for digitally low-resource Indigenous languages is often hindered by extreme data scarcity, prompting reliance on extractive web-scraping.
By Alexander Chulzhanov, Soeren Eberhardt, Arjun Mukherjee
arXiv:2607. 24276v1 Announce Type: cross Abstract: Large language models (LLMs) process text through subword tokenizers rather than directly reading characters or words.
By Priyansh Srivastava