@inproceedings{dutta-etal-2025-graft,
title = "{GRAFT}: A Graph-based Flow-aware Agentic Framework for Document-level Machine Translation",
author = "Dutta, Himanshu and
Manchanda, Sunny and
Bapat, Prakhar and
Gurjar, Meva Ram and
Bhattacharyya, Pushpak",
editor = "Potdar, Saloni and
Rojas-Barahona, Lina and
Montella, Sebastien",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = nov,
year = "2025",
address = "Suzhou (China)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-industry.166/",
pages = "2405--2428",
ISBN = "979-8-89176-333-3",
abstract = "Enterprises, public organizations, and localization providers increasingly rely on Document-level Machine Translation (DocMT) to process contracts, reports, manuals, and multimedia transcripts across languages. However, existing MT systems often struggle to handle discourse-level phenomena such as pronoun resolution, lexical cohesion, and ellipsis, resulting in inconsistent or incoherent translations. We propose **GRAFT**, a modular graph-based DocMT framework that leverages Large Language Model (LLM) agents to segment documents into discourse units, infer inter-discourse dependencies, extract structured memory, and generate context-aware translations. GRAFT transforms documents into directed acyclic graphs (DAGs) to explicitly model translation flow and discourse structure. Experiments across eight language directions and six domains show GRAFT outperforms commercial systems (e.g., $\texttt{Google Translate}$) and closed LLMs (e.g., $\texttt{GPT-4}$) by an average of $\textbf{2.8}$ d-BLEU, and improves terminology consistency and discourse handling. GRAFT supports deployment with open-source LLMs (e.g., LLaMA, Qwen), making it cost-effective and privacy-preserving. These results position GRAFT as a robust solution for scalable, document-level translation in real-world applications."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dutta-etal-2025-graft">
<titleInfo>
<title>GRAFT: A Graph-based Flow-aware Agentic Framework for Document-level Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Himanshu</namePart>
<namePart type="family">Dutta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunny</namePart>
<namePart type="family">Manchanda</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Prakhar</namePart>
<namePart type="family">Bapat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meva</namePart>
<namePart type="given">Ram</namePart>
<namePart type="family">Gurjar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Saloni</namePart>
<namePart type="family">Potdar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lina</namePart>
<namePart type="family">Rojas-Barahona</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sebastien</namePart>
<namePart type="family">Montella</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou (China)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-333-3</identifier>
</relatedItem>
<abstract>Enterprises, public organizations, and localization providers increasingly rely on Document-level Machine Translation (DocMT) to process contracts, reports, manuals, and multimedia transcripts across languages. However, existing MT systems often struggle to handle discourse-level phenomena such as pronoun resolution, lexical cohesion, and ellipsis, resulting in inconsistent or incoherent translations. We propose **GRAFT**, a modular graph-based DocMT framework that leverages Large Language Model (LLM) agents to segment documents into discourse units, infer inter-discourse dependencies, extract structured memory, and generate context-aware translations. GRAFT transforms documents into directed acyclic graphs (DAGs) to explicitly model translation flow and discourse structure. Experiments across eight language directions and six domains show GRAFT outperforms commercial systems (e.g., Google Translate) and closed LLMs (e.g., GPT-4) by an average of 2.8 d-BLEU, and improves terminology consistency and discourse handling. GRAFT supports deployment with open-source LLMs (e.g., LLaMA, Qwen), making it cost-effective and privacy-preserving. These results position GRAFT as a robust solution for scalable, document-level translation in real-world applications.</abstract>
<identifier type="citekey">dutta-etal-2025-graft</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-industry.166/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>2405</start>
<end>2428</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T GRAFT: A Graph-based Flow-aware Agentic Framework for Document-level Machine Translation
%A Dutta, Himanshu
%A Manchanda, Sunny
%A Bapat, Prakhar
%A Gurjar, Meva Ram
%A Bhattacharyya, Pushpak
%Y Potdar, Saloni
%Y Rojas-Barahona, Lina
%Y Montella, Sebastien
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou (China)
%@ 979-8-89176-333-3
%F dutta-etal-2025-graft
%X Enterprises, public organizations, and localization providers increasingly rely on Document-level Machine Translation (DocMT) to process contracts, reports, manuals, and multimedia transcripts across languages. However, existing MT systems often struggle to handle discourse-level phenomena such as pronoun resolution, lexical cohesion, and ellipsis, resulting in inconsistent or incoherent translations. We propose **GRAFT**, a modular graph-based DocMT framework that leverages Large Language Model (LLM) agents to segment documents into discourse units, infer inter-discourse dependencies, extract structured memory, and generate context-aware translations. GRAFT transforms documents into directed acyclic graphs (DAGs) to explicitly model translation flow and discourse structure. Experiments across eight language directions and six domains show GRAFT outperforms commercial systems (e.g., Google Translate) and closed LLMs (e.g., GPT-4) by an average of 2.8 d-BLEU, and improves terminology consistency and discourse handling. GRAFT supports deployment with open-source LLMs (e.g., LLaMA, Qwen), making it cost-effective and privacy-preserving. These results position GRAFT as a robust solution for scalable, document-level translation in real-world applications.
%U https://aclanthology.org/2025.emnlp-industry.166/
%P 2405-2428
Markdown (Informal)
[GRAFT: A Graph-based Flow-aware Agentic Framework for Document-level Machine Translation](https://aclanthology.org/2025.emnlp-industry.166/) (Dutta et al., EMNLP 2025)
ACL