@inproceedings{chitale-etal-2026-updesh,
title = "{UPDESH}: Synthesizing Grounded Instruction Tuning Data for 13 {I}ndic Languages",
author = "Chitale, Pranjal A and
Gumma, Varun and
Ahuja, Sanchit and
Kodali, Prashant and
Uppadhyay, Manan and
Sudharsan, Deepthi and
Sitaram, Sunayana",
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.1763/",
pages = "37997--38041",
ISBN = "979-8-89176-390-6",
abstract = "Developing culturally grounded multilingual AI systems remains challenging, particularly for low-resource languages. While synthetic data offers promise, its effectiveness in multilingual and multicultural contexts is underexplored. We investigate bottom-up synthetic data generation using large open-source LLMs ({\ensuremath{>}}= 235B parameters) grounded in language-specific Wikipedia content, complementing dominant top-down translation-based approaches from English. We introduce Updesh, a high-quality large-scale synthetic instruction-following dataset comprising 9.5M data points across 13 Indian languages and English, encompassing diverse reasoning and generative tasks emphasizing on enhancing long-context and multi-turn capabilities while improving alignment with Indian cultural contexts. Comprehensive evaluation using automated metrics and 10K human assessments confirms high data quality. Downstream evaluations performed by fine-tuning models on various datasets and assessing performance across 13 diverse multilingual datasets and model comparative evaluations, demonstrate that models trained on Updesh consistently obtain significant improvements on NLG tasks and remain competitive on NLU tasks. Improvements are most pronounced for low and medium-resource languages, effectively narrowing performance gaps with high-resource languages. Our findings provide empirical evidence that effective multilingual AI development requires multi-faceted, culturally grounded data curation strategies beyond translation-based approaches."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chitale-etal-2026-updesh">
<titleInfo>
<title>UPDESH: Synthesizing Grounded Instruction Tuning Data for 13 Indic Languages</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pranjal</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Chitale</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Varun</namePart>
<namePart type="family">Gumma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sanchit</namePart>
<namePart type="family">Ahuja</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Prashant</namePart>
<namePart type="family">Kodali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manan</namePart>
<namePart type="family">Uppadhyay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Deepthi</namePart>
<namePart type="family">Sudharsan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunayana</namePart>
<namePart type="family">Sitaram</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>Developing culturally grounded multilingual AI systems remains challenging, particularly for low-resource languages. While synthetic data offers promise, its effectiveness in multilingual and multicultural contexts is underexplored. We investigate bottom-up synthetic data generation using large open-source LLMs (\ensuremath>= 235B parameters) grounded in language-specific Wikipedia content, complementing dominant top-down translation-based approaches from English. We introduce Updesh, a high-quality large-scale synthetic instruction-following dataset comprising 9.5M data points across 13 Indian languages and English, encompassing diverse reasoning and generative tasks emphasizing on enhancing long-context and multi-turn capabilities while improving alignment with Indian cultural contexts. Comprehensive evaluation using automated metrics and 10K human assessments confirms high data quality. Downstream evaluations performed by fine-tuning models on various datasets and assessing performance across 13 diverse multilingual datasets and model comparative evaluations, demonstrate that models trained on Updesh consistently obtain significant improvements on NLG tasks and remain competitive on NLU tasks. Improvements are most pronounced for low and medium-resource languages, effectively narrowing performance gaps with high-resource languages. Our findings provide empirical evidence that effective multilingual AI development requires multi-faceted, culturally grounded data curation strategies beyond translation-based approaches.</abstract>
<identifier type="citekey">chitale-etal-2026-updesh</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.1763/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>37997</start>
<end>38041</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T UPDESH: Synthesizing Grounded Instruction Tuning Data for 13 Indic Languages
%A Chitale, Pranjal A.
%A Gumma, Varun
%A Ahuja, Sanchit
%A Kodali, Prashant
%A Uppadhyay, Manan
%A Sudharsan, Deepthi
%A Sitaram, Sunayana
%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 chitale-etal-2026-updesh
%X Developing culturally grounded multilingual AI systems remains challenging, particularly for low-resource languages. While synthetic data offers promise, its effectiveness in multilingual and multicultural contexts is underexplored. We investigate bottom-up synthetic data generation using large open-source LLMs (\ensuremath>= 235B parameters) grounded in language-specific Wikipedia content, complementing dominant top-down translation-based approaches from English. We introduce Updesh, a high-quality large-scale synthetic instruction-following dataset comprising 9.5M data points across 13 Indian languages and English, encompassing diverse reasoning and generative tasks emphasizing on enhancing long-context and multi-turn capabilities while improving alignment with Indian cultural contexts. Comprehensive evaluation using automated metrics and 10K human assessments confirms high data quality. Downstream evaluations performed by fine-tuning models on various datasets and assessing performance across 13 diverse multilingual datasets and model comparative evaluations, demonstrate that models trained on Updesh consistently obtain significant improvements on NLG tasks and remain competitive on NLU tasks. Improvements are most pronounced for low and medium-resource languages, effectively narrowing performance gaps with high-resource languages. Our findings provide empirical evidence that effective multilingual AI development requires multi-faceted, culturally grounded data curation strategies beyond translation-based approaches.
%U https://aclanthology.org/2026.acl-long.1763/
%P 37997-38041
Markdown (Informal)
[UPDESH: Synthesizing Grounded Instruction Tuning Data for 13 Indic Languages](https://aclanthology.org/2026.acl-long.1763/) (Chitale et al., ACL 2026)
ACL
- Pranjal A Chitale, Varun Gumma, Sanchit Ahuja, Prashant Kodali, Manan Uppadhyay, Deepthi Sudharsan, and Sunayana Sitaram. 2026. UPDESH: Synthesizing Grounded Instruction Tuning Data for 13 Indic Languages. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 37997–38041, San Diego, California, United States. Association for Computational Linguistics.