@inproceedings{niyogi-etal-2026-paramanu,
title = "Paramanu: Compact and Competitive Monolingual Language Models for Low-Resource Morphologically Rich {I}ndian Languages",
author = "Niyogi, Mitodru and
Gaussier, Eric and
Bhattacharya, Arnab",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.1922/",
pages = "41431--41458",
ISBN = "979-8-89176-390-6",
abstract = "Multilingual large language models (LLMs) are expensive to pretrain and often suffer from imbalances across languages and datasets, English-centric bias, tokenizer oversegmentation for morphologically rich low-resource languages, and the curse of multilinguality. We introduce PARAMANU, a family of Indian language-only autoregressive language models trained from scratch on open-source language-specific data for the five most spoken Indian languages: Bangla (Bengali), Hindi, Marathi, Tamil, and Telugu. All models are designed for affordability and are trained on a single GPU with a budget under $1,000, allowing under-resourced researchers to build competitive language models. To address low-resource challenges, we develop morphology-aligned, low-fertility tokenizers, and propose an interpolation-based method for token position indices in RoPE based scaling to train longer sequences efficiently. We also create instruction-tuning datasets in Bangla that are then translated to the other four languages. Despite their small size (108M-367M parameters), Paramanu achieves a strong performance-efficiency tradeoff and outperforms most larger multilingual models up to 8B across all five languages. The models and datasets are available at: https://huggingface.co/collections/mitodru/paramanu.$"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="niyogi-etal-2026-paramanu">
<titleInfo>
<title>Paramanu: Compact and Competitive Monolingual Language Models for Low-Resource Morphologically Rich Indian Languages</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mitodru</namePart>
<namePart type="family">Niyogi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eric</namePart>
<namePart type="family">Gaussier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arnab</namePart>
<namePart type="family">Bhattacharya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Multilingual large language models (LLMs) are expensive to pretrain and often suffer from imbalances across languages and datasets, English-centric bias, tokenizer oversegmentation for morphologically rich low-resource languages, and the curse of multilinguality. We introduce PARAMANU, a family of Indian language-only autoregressive language models trained from scratch on open-source language-specific data for the five most spoken Indian languages: Bangla (Bengali), Hindi, Marathi, Tamil, and Telugu. All models are designed for affordability and are trained on a single GPU with a budget under 1,000, allowing under-resourced researchers to build competitive language models. To address low-resource challenges, we develop morphology-aligned, low-fertility tokenizers, and propose an interpolation-based method for token position indices in RoPE based scaling to train longer sequences efficiently. We also create instruction-tuning datasets in Bangla that are then translated to the other four languages. Despite their small size (108M-367M parameters), Paramanu achieves a strong performance-efficiency tradeoff and outperforms most larger multilingual models up to 8B across all five languages. The models and datasets are available at: https://huggingface.co/collections/mitodru/paramanu.</abstract>
<identifier type="citekey">niyogi-etal-2026-paramanu</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.1922/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>41431</start>
<end>41458</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Paramanu: Compact and Competitive Monolingual Language Models for Low-Resource Morphologically Rich Indian Languages
%A Niyogi, Mitodru
%A Gaussier, Eric
%A Bhattacharya, Arnab
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F niyogi-etal-2026-paramanu
%X Multilingual large language models (LLMs) are expensive to pretrain and often suffer from imbalances across languages and datasets, English-centric bias, tokenizer oversegmentation for morphologically rich low-resource languages, and the curse of multilinguality. We introduce PARAMANU, a family of Indian language-only autoregressive language models trained from scratch on open-source language-specific data for the five most spoken Indian languages: Bangla (Bengali), Hindi, Marathi, Tamil, and Telugu. All models are designed for affordability and are trained on a single GPU with a budget under 1,000, allowing under-resourced researchers to build competitive language models. To address low-resource challenges, we develop morphology-aligned, low-fertility tokenizers, and propose an interpolation-based method for token position indices in RoPE based scaling to train longer sequences efficiently. We also create instruction-tuning datasets in Bangla that are then translated to the other four languages. Despite their small size (108M-367M parameters), Paramanu achieves a strong performance-efficiency tradeoff and outperforms most larger multilingual models up to 8B across all five languages. The models and datasets are available at: https://huggingface.co/collections/mitodru/paramanu.
%U https://aclanthology.org/2026.acl-long.1922/
%P 41431-41458
Markdown (Informal)
[Paramanu: Compact and Competitive Monolingual Language Models for Low-Resource Morphologically Rich Indian Languages](https://aclanthology.org/2026.acl-long.1922/) (Niyogi et al., ACL 2026)
ACL